CN107301234A - A kind of voice and word interactive system based on regulation and control atypia data - Google Patents

Abstract

The invention relates to a voice and text interactive system based on regulation and control of atypical data, which solves the technical problems of cumbersome operation and low accuracy. By adopting a main system login module and a speech and text recognition system, the atypical data dictionary construction module is controlled. And an alarm output module; the voice and text recognition system includes a voice and text configuration module and a voice and text recognition module; the voice and text configuration module is used to configure input data, and the input data includes voice and text; the voice and text recognition module is used To formulate an expert strategy, maximize the matching and comparison source and input data, compare and analyze the matching result with the expert strategy, and obtain the analysis result; the regulation and control atypical data dictionary construction module is used to establish a regulation and control atypical data dictionary, regulate The dictionary is used as a technical solution for comparison sources, which solves this problem well and can be used in smart grids.

Description

A Speech and Text Interaction System Based on Controlling Atypical Data

technical field

The invention relates to the field of smart grids, in particular to a voice and text interactive system based on regulating atypical data.

Background technique

In recent years, my country's electric power industry has been in a period of vigorous development and has achieved remarkable achievements. In order to promote the sustainable development of the power grid, the field of power grid dispatching automation is facing technological innovation, and intelligent dispatching has become a new direction of its development. Regulatory instructions are an important task in the daily production of power grid dispatching, and are also an important part of the work of operating departments such as power plants and substations. The correctness of regulatory instructions is directly related to the correctness of operations, and is related to equipment, personal and grid safety. Electrical misoperation may cause equipment damage, personal casualties, and even cause grid oscillation and disintegration, leading to serious consequences such as large-scale power outages.

The ever-expanding power system requires more monitoring information, and the daily work intensity of power grid dispatchers increases accordingly, which weakens the ability to deal with unexpected accidents. At present, the control operation, control management and atypical data of power grid dispatching need to be judged, analyzed and compared manually, which has the problems of cumbersome operation and low accuracy. Therefore, it is necessary to provide a voice and text interaction system based on regulating atypical data.

Contents of the invention

The technical problem to be solved by the invention is the technical problem of cumbersome operation and low accuracy existing in the prior art. A new voice and text interactive system based on regulating atypical data is provided, and the voice and text interactive system based on regulating atypical data has the characteristics of simple operation and high accuracy.

In order to solve the above technical problems, the technical scheme adopted is as follows:

A voice and text interactive system based on regulation and control of atypical data, the voice and text interaction system includes a main system login module, a voice and text recognition system, a regulation and control atypical data dictionary construction module and an alarm output module; the main system login module It is used to verify the user identity of the voice and text interactive system; the voice and text recognition system includes a voice and text configuration module and a voice and text recognition module; the voice and text configuration module is used to configure input data, and the input data includes voice and text The voice and text recognition module is used to formulate expert strategies, maximize the matching comparison source and input data, compare and analyze the matching results with expert strategies, and obtain analysis results; the regulation and control atypical data dictionary construction module is used to establish regulation and control The atypical data dictionary is adjusted and used as a comparison source; the alarm output module is used to complete information interaction with the voice and text interaction system users, and transmit the analysis results to the voice and text interaction system users.

The working principle of the present invention: the present invention can provide dispatchers with more dispatching information by introducing the human-computer interaction technology of voice and text into the intelligent dispatching system, and complete the information exchange between the dispatcher and the dispatching system through voice. Specifically, users of the voice and text interaction system can regulate the construction and configuration of atypical data dictionaries. Import and control atypical data dictionaries, and enter the stage of listening to the voice and text of dispatch commands. When information such as voice or text is input, the information includes dispatch logs, operation tickets, command voices, and application forms for planned outages, etc. Match the structure to the maximum extent, analyze and check the expert strategy formulated by the system and the current power grid operation status, and send the analysis results to the system background in the format of structured data, and the system background will feed back the results to the interactive program after query and analysis , the result will be transmitted to the user of the voice and text interaction system in the form of an alarm.

In the above solution, for optimization, further, the main system login module includes a permission module, and the permission module includes an identity data input unit, an identity database, and an identity data matching module;

The identity data input unit includes a data collection device for collecting identity information of users of the voice and text interactive system, and the identity information includes fingerprint information, pupil information, voice information or text password information;

The identity database is used for pre-storing identity data corresponding to authority classification;

The identity data matching module is used to match the identity data with the identity database.

Further, the identity data matching module includes an entity conversion unit, a data reflection unit, a matching verification unit and a custom characteristic definition unit;

The entity conversion unit is used to define the identity information collected by the identity data input unit as an entity;

The self-defined characteristic unit encapsulates verification logic as a self-defined characteristic;

The data reflection unit obtains field values and corresponding custom characteristic values through reflection according to entities;

The matching verification unit is used to verify the field value and the custom characteristic value.

Further, the alarm output module includes an alarm module and an output module, the alarm module is used to notify the user of the voice and text interaction system of the abnormal state; the output module is used to output all information of the voice and text interaction system to the voice and text interaction system users.

Further, the speech and text recognition module includes a connection device, which is used to connect the speech and text interaction system and the speech and text interaction system user, and the connection device includes a storage medium and a processor, and the storage medium stores two-way processing Program, two-way processing program comprises one-way and reverse program, and described processor is used for carrying out two-way processing program, to complete following steps:

Judging the request flow direction, the request flows from the voice and text interaction system user to the voice and text interaction system, and executes the one-way procedure, and the request flow direction is from the voice and text interaction system to the voice and text interaction system user, and executes the reverse procedure;

The one-way program only visits the voice and text interaction system once according to the voice and text interaction system user’s request; the reverse program sends the voice and text interaction system request to the voice and text interaction system user, and uses it according to the voice and text interaction system The respondent accesses the voice and text interaction system only once.

Further, the one-way procedure includes that the connecting device filters repeated requests from users of the voice and text interaction system, and sends the request received for the first time to the voice and text interaction system;

(A) At time t, receive a request from the user of the voice and text interaction system, check and judge that the request received at time t is the first time received by the connection device, execute step (B), the request received at time t is not the first time received by the connection device, execute step (D);

(B) Send the request in step (A) to the voice and text interaction system, and add the request received at time t as a historical request to the request sequence;

(C) The voice and text interaction system returns instructions, and the connection device receives the return result of the voice and text interaction system, and adds the returned result to the return sequence, which corresponds to the request sequence in step (B);

(D) The connection device processes the historical request in the search request sequence that is the same as the request received at time t, and searches for the return result corresponding to the historical request in the return queue;

(E) Send the return result in the return queue to the voice and text interaction system node corresponding to the request at time t.

Further, the reverse procedure includes that the connection device receives the request from the voice and text interaction system, sends the request to the user of the voice and text interaction system, and after the connection device receives the response from the voice and text interaction system user, filters Response from the voice and text interaction system user, returning the first received response to the voice and text interaction system;

(a) The connection device receives the request from the voice and text interactive system, sends the request to the voice and text interactive system user, and adds the request to the request queue;

(b) The connection device receives the response result from the user of the voice and text interaction system at time t, and consults the response result corresponding to the request in step (a) that is determined to be received for the first time, executes (c), and consults the determination response result If it is not the response result corresponding to the request of step (a) received for the first time, the connection device adds the response result received at time t to the return queue, and the return queue corresponds to the request queue;

(c) The connection device returns the response result received at time t to the voice and text interaction system as a request.

Further, the input data includes scheduling logs, operation tickets, command voices and planned outage application forms.

Further, the voice and text interaction system is used for smart grid.

The invention adds a connecting device to the voice and character recognition module to connect with the alarm output module, and is used to deal with the situation that multiple users send requests to the interactive system at the same time, or the interactive system issues commands to multiple users at the same time.

The data of the voice and text interaction system is open to all nodes in the network relying on the decentralized P2P network, and users of the voice and text interaction system can participate in power grid scheduling through the open interface. The voice and text interaction system relies on the asymmetric encryption technology of modern cryptography to ensure the security of transaction information, and relies on a consensus process called workload proof mechanism to ensure that the data of the voice and text interaction system cannot be tampered with. The data of the voice and text interaction system is generated, disseminated, and stored in a distributed manner in the P2P network, and each node stores a complete data record. Therefore, if some nodes fail, as long as there is a normal operating node, it will not affect the update of block data.

Beneficial effects of the present invention:

Effect 1. By introducing voice and text human-computer interaction technology into the intelligent dispatching system, more dispatching information can be provided to dispatchers, and the information exchange between dispatchers and the dispatching system can be completed through voice, thereby significantly reducing the dispatcher's workload. Intensity of work;

The second effect is to effectively improve the efficiency of power dispatching;

Effect 3, the interaction effect is accurate and convenient;

Effect 4: By using the connecting device, one-to-many and many-to-one can be quickly processed without error.

Description of drawings

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

Fig. 1 is a schematic diagram of the workflow in Example 1.

Fig. 2 is a schematic diagram of the connection of the connection device in Embodiment 1.

Fig. 3 is a system schematic diagram of the voice and text interaction system in Embodiment 1.

detailed description

In order to make the object, technical solution and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the examples. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

Example 1

This embodiment provides a voice and text interactive system based on regulation and control of atypical data. The voice and text interaction system can be used in smart grids, as shown in Figure 3. The voice and text interaction system includes a main system login module, a voice and text recognition system, Regulating the atypical data dictionary construction module and the alarm output module; the main system login module is used to verify the user identity of the voice and text interactive system; the voice and text recognition system includes a voice and text configuration module and a voice and text recognition module; The voice and text configuration module is used to configure input data, and the input data includes voice and text; the voice and text recognition module is used to formulate expert strategies, maximize matching and comparison sources and input data, and compare and analyze matching results with expert strategies, Obtain the analysis results; the regulation and control atypical data dictionary construction module is used to establish the regulation and control atypical data dictionary, and the regulation and control atypical data dictionary is used as a comparison source; the described alarm output module is used to complete the interaction with the voice and text interaction system users Information interaction, which transmits the analysis results to users of the voice and text interaction system.

Workflow of this embodiment: users of the voice and text interaction system can control the construction and configuration of atypical data dictionaries. Import and control atypical data dictionaries, and enter the stage of listening to the voice and text of dispatch commands. When information such as voice or text is input, the information includes dispatch logs, operation tickets, command voices, and application forms for planned outages, etc. Match the structure to the maximum extent, analyze and check the expert strategy formulated by the system and the current power grid operation status, and send the analysis results to the system background in the format of structured data, and the system background will feed back the results to the interactive program after query and analysis , the result will be transmitted to the user of the voice and text interaction system in the form of an alarm.

The maximum matching workflow in the Speech and Text Recognition module includes:

(A) identification system initialization, including initializing the COM platform;

(B) importing input data;

(C) Create a recognition engine object, the recognition engine object is a speech recognition engine object and a text recognition engine object, start the recognition engine; determine the working state of the recognition engine, and the working state of the recognition engine is the working state, then perform step (D); the recognition engine works If the state is dormant, execute step (B);

(D) Establish a message notification mechanism, including sending custom information to activate the information function, and loading and activating grammar rules according to the information function;

(E) The grammatical rules include keywords and sentence structures, and match the speech in the input data with the keywords and sentence structures in the grammatical rules to the maximum; determine the matching result, if the matching is successful, execute step (F); if the matching fails , execute step (B);

(F) Defining the matching result as the speech recognition output result, outputting the speech recognition output result in text form, and uninstalling the COM platform.

Specifically, the main system login module includes a permission module, and the permission module includes an identity data input unit, an identity database, and an identity data matching module; the identity data input unit includes a data collection device for collecting voice and text interaction The identity information of the system user, the identity information includes fingerprint information, pupil information, voice information or text password information; the identity database is used to pre-store the identity data corresponding to the authority classification; the identity data matching module is used to match the identity data with the identity database.

Preferably, in order to complete identity data matching at high speed, the identity data matching module includes an entity conversion unit, a data reflection unit, a matching verification unit and a custom characteristic definition unit;

The entity conversion unit is used to define the identity information collected by the identity data input unit as an entity;

The self-defined characteristic unit encapsulates verification logic as a self-defined characteristic;

The data reflection unit obtains field values and corresponding custom characteristic values through reflection according to entities;

The matching verification unit is used to verify the field value and the custom characteristic value.

Through custom properties, use reflection to get all the field values of the entity, and the corresponding custom property values of the fields; compare the field values with the custom property values, specifically, through the custom property values, you can know whether the field is required, and the field Whether the value is empty, whether the field length is too long, whether the field value type meets the requirements, etc., finally verify the validity of the data. And no matter how many fields the entity has, it is no longer necessary to manually write a lot of code to validate the data. Encapsulate the verification logic into the feature, and add this feature to any field that needs to be verified. When verifying data, you only need to obtain the characteristic value through reflection to do the verification work.

Specifically, the alarm output module includes an alarm module and an output module, the alarm module is used to notify the user of the voice and text interaction system of the abnormal state; the output module is used to output all information of the voice and text interaction system to the voice and text interaction system users.

In order to handle the complex interaction requirements of one-to-many and many-to-one simply and error-free. Preferably, as shown in Figure 2, the speech and text recognition module includes a connection device, the connection device is used to connect the speech and text interaction system and the speech and text interaction system user, and the connection device includes a storage medium, a processor, a storage medium Two-way processing programs are stored, and the two-way processing programs include one-way and reverse programs, and the processor is used to execute the two-way processing programs to complete the following steps:

Judging the request flow direction, the request flows from the voice and text interaction system user to the voice and text interaction system, and executes the one-way procedure, and the request flow direction is from the voice and text interaction system to the voice and text interaction system user, and executes the reverse procedure;

The one-way program only visits the voice and text interaction system once according to the voice and text interaction system user’s request; the reverse program sends the voice and text interaction system request to the voice and text interaction system user, and uses it according to the voice and text interaction system The respondent accesses the voice and text interaction system only once.

Specifically, the one-way procedure includes that the connecting device filters repeated requests from users of the voice and text interaction system, and sends the request received for the first time to the voice and text interaction system;

(A) At time t, receive a request from the user of the voice and text interaction system, check and judge that the request received at time t is the first time received by the connection device, execute step (B), the request received at time t is not the first time received by the connection device, execute step (D);

(B) Send the request in step (A) to the voice and text interaction system, and add the request received at time t as a historical request to the request sequence;

(C) The voice and text interaction system returns instructions, and the connection device receives the return result of the voice and text interaction system, and adds the returned result to the return sequence, which corresponds to the request sequence in step (B);

(D) The connection device processes the historical request in the search request sequence that is the same as the request received at time t, and searches for the return result corresponding to the historical request in the return queue;

(E) Send the return result in the return queue to the voice and text interaction system node corresponding to the request at time t.

Specifically, the reverse procedure includes that the connection device receives the request from the voice and text interaction system, sends the request to the voice and text interaction system user, and after the connection device receives the voice and text interaction system user's response, filters Response from the voice and text interaction system user, returning the first received response to the voice and text interaction system;

(a) The connection device receives the request from the voice and text interactive system, sends the request to the voice and text interactive system user, and adds the request to the request queue;

(b) The connection device receives the response result from the user of the voice and text interaction system at time t, and consults the response result corresponding to the request in step (a) that is determined to be received for the first time, executes (c), and consults the determination response result If it is not the response result corresponding to the request of step (a) received for the first time, the connection device adds the response result received at time t to the return queue, and the return queue corresponds to the request queue;

(c) The connection device returns the response result received at time t to the voice and text interaction system as a request.

Specifically, the input data includes scheduling logs, operation tickets, command voices and planned outage application forms.

Although the illustrative specific embodiments of the present invention have been described above, so that those skilled in the art can understand the present invention, the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, as long as each As long as the changes are within the spirit and scope of the present invention defined and determined by the appended claims, all inventions and creations utilizing the concept of the present invention are included in the protection list.

Claims (9)

1. a kind of voice and word interactive system based on regulation and control atypia data, it is characterised in that：The voice and pen friends Mutual system includes main system login module, and language and characters identifying system, regulation and control atypia data dictionary builds module and alarm Output module；

The main system login module is used to verify voice and word interaction system user identity；

The language and characters identifying system includes voice and word configuration module and voice and Text region module；

Voice and the word configuration module is used to configure input data, and input data includes voice and word；

Voice and the Text region module is used to formulate expert tactics, maximizes matching contrast source and input data, will match As a result with expert tactics comparative analysis, analysis result is drawn；

The regulation and control atypia data dictionary, which builds module, to be used to set up regulation and control atypia data dictionary, regulates and controls atypia data word Allusion quotation is used for source as a comparison；

The alarm output module is used to interact system user completion information exchange with voice and word, and analysis result is transmitted To voice and word interaction system user.

2. voice and word interactive system according to claim 1 based on regulation and control atypia data, it is characterised in that：Institute State includes authority module including main system login module, and the authority module includes identity data input block, identity database, Identification data matches module；

The identity data input block includes data acquisition device, and the body of system user is interacted for gathering voice and word Part information, identity information includes finger print information, pupil information, acoustic information or word encrypted message；

The identity database is used to prestore the identity data corresponding to authority classification；

The identification data matches module is used to match identity data and identity database.

3. voice and word interactive system according to claim 2 based on regulation and control atypia data, it is characterised in that：

The identification data matches module includes entity converting unit, data reflector element, matching authentication unit and self-defined spy Property definition unit；

The identity information that the entity converting unit is used to gather identity data input block is defined as entity；

Verifying logic is encapsulated as custom properties by the custom properties unit；

The data reflector element obtains field value, corresponding custom properties value according to entity by reflecting；

The matching authentication unit is used to be verified field value with custom properties value.

4. voice and word interactive system according to claim 1 based on regulation and control atypia data, it is characterised in that：Institute Stating alarm output module includes alarm module and output module, and the alarm module is used to abnormality notifying voice and word Interactive system user；The output module is used to export voice and all information of word interactive system to voice and pen friends Mutual system user.

5. voice and word interactive system according to claim 1 based on regulation and control atypia data, it is characterised in that：Institute Predicate sound and Text region module include attachment means, and attachment means are used to connect voice and word interactive system and voice and text Word interactive system user, the attachment means include storage medium, processor, and storage medium is stored with two-way processing routine, Two-way processing routine includes unidirectional and reversal procedure, and the processor is used to perform two-way processing routine, to complete the following steps：

Judge request flow direction, request flows to voice and word interacts system user to voice and word interactive system, performs list To program, request flow direction is that voice and word interactive system interact system user to voice and word, performs reversal procedure；

Unidirectional program one way accesses voice and word interactive system once according only to voice and word interaction system user request； Voice and word interaction system request are sent to reversal procedure into voice and word interacts system user, and according to voice and text The response of word interactive system user only accesses voice and word interactive system once.

6. voice and word interactive system according to claim 5 based on regulation and control atypia data, it is characterised in that：

The unidirectional program includes, and attachment means filtering interacts the repetitive requests of system user from voice and word, by head The secondary request received is sent to voice and word interactive system；

（A）T, receives the request that system user is interacted from voice and word, consults and judges that the request that t is received is Attachment means are received first, perform step（B）, the request that t is received is not that attachment means are received first, performs step（D）；

（B）By step（A）In request be issued to voice and word interactive system, and the request that t is received is used as history Request adds request sequence；

（C）Voice and word interactive system return instruction, attachment means receive the returning result of voice and word interactive system, And returning result is added into return sequence, return to sequence and step（B）In request sequence it is corresponding；

（D）The request identical historical requests received in attachment means processing search request sequence with t, and search return team Returning result corresponding with historical requests in row；

（E）The returning result returned in queue is sent to t and asks corresponding voice and word interaction system node.

7. voice and word interactive system according to claim 5 based on regulation and control atypia data, it is characterised in that：

The reversal procedure includes, and attachment means receive the request from voice and word interactive system, send the requests to language Sound and word interaction system user, attachment means are received after the response of voice and word interaction system user, and filtering comes The response of system user is interacted from voice and word, the response received first is returned into voice and word interactive system；

（a）Attachment means receive the request from voice and word interactive system, transmit the request to voice and word interaction system United user, and request is added into request queue；

（b）Attachment means t receives the response results in voice and word interaction system user, consults and judges to ring Should result for receive first the step of（a）The corresponding response results of request, perform（c）, consult headed by judging response results not The step of secondary reception（a）The corresponding response results of request, the response results that attachment means receive t, which are added, returns to team Row, return to queue corresponding with request queue；

（c）The response results that attachment means receive t return to voice and word interactive system as request.

8. voice and word interactive system according to claim 1 based on regulation and control atypia data, it is characterised in that：Institute Stating input data includes dispatching log, operation order, issue an order voice and planned outage application form.

9. voice and word interactive system according to claim 1 based on regulation and control atypia data, it is characterised in that：Institute Predicate sound and word interactive system are used for intelligent grid.