CN102682560B - Device for assessing level of fire interlock alarming in ship cabin - Google Patents

Abstract

The invention provides a method and device for assessing the fire chain alarm level of a ship cabin. 1) The centralized fire alarm controller collects the fire information obtained by the temperature and smoke detectors in the cabins adjacent to the fire compartment; 2) The centralized fire alarm controller reads the fire hazard level information of the cabin; 3) The temperature sensor 1. The fire information obtained by the smoke detector and the cabin fire hazard level information are normalized and preprocessed; 4) The processed data is sent to the fuzzy reasoning module for fuzzy reasoning, so as to obtain the evaluation result of the cabin fire chain alarm level. The present invention analyzes and processes the data of different types of detector information in each cabin and the fire hazard level information of the cabin itself through the fuzzy reasoning system without adding redundant detection equipment and costs, and quickly and accurately gives the alarm level information of each cabin. Strive to achieve the accuracy, advanceness and comprehensiveness of cabin chain alarms, and reduce major fire accidents caused by omission of cabin fire information.

Description

A fire chain alarm level evaluation device for ship cabins

technical field

The present invention relates to an automatic fire alarm method, in particular to an intelligent fire alarm method, in particular to a fuzzy control-based fire chain alarm level evaluation method and device for ship cabins. the

Background technique

With the development of science and technology, countries around the world are vigorously developing the maritime industry. As an important tool for maritime transportation, ships have become the focus of attention. Due to the complex environment of modern ships, the cabin space on board is small, the mechanical equipment is tightly placed, the gas and liquid transportation and circuit lines are complicated, so when a fire occurs in a cabin in a certain area, the adjacent cabins will inevitably be implicated. If the fire is not discovered in time , as the fire intensifies, it will spread to several cabins through walls or interconnected ventilation pipes, oil and gas transportation pipes, cables, etc., causing chain fires, thus posing a huge threat to the safety of the main body of the ship and personnel. The extent and possible losses will be far greater than that of general building fires. the

At present, my country's understanding of how to effectively control and deal with ship fires is not comprehensive enough, it is only at the stage of macro and experience, and it lacks a complete ship fire safety assessment system. In recent years at home and abroad, the research on intelligent fire alarm system tends to be carried out. For example, the intelligent fire detector is designed by using the control method combined with neural network and fuzzy system and the fusion technology of multi-alarm parameter information. Learning and adaptability, as well as certain reliability, can effectively reduce the false alarm rate of the detector. In addition, artificial intelligence and interactive technologies are also adopted by more and more intelligent fire alarm systems, which have become the frontier and core technologies of fire alarm systems. However, today's intelligent fire alarm system only focuses on the accurate judgment of the fire by the detector, reduces the frequency of false alarms and omissions, improves the system response sensitivity, and starts the fire extinguishing facilities quickly, etc., while ignoring the fire alarm system. The importance of "mainly" and supplemented by "elimination". The cabin is an important part of the ship. When a fire occurs in a certain cabin, according to the characteristics of the cabin fire, the fire in the cabin will spread rapidly, which may cause chain fires, and the consequences will be disastrous. Reliable and accurate interlock alarm technology for the cabin is very important. the

When a fire breaks out in a cabin, the alarm system should not only obtain the exact location of the cabin, fire parameters, alarm level and other information, but also collect the environmental information and fire information of the adjacent cabins around it, and comprehensively judge and determine the fire situation of these cabins. Development and alarm level information. This can not only make up for the functional defects of the ship's fire alarm system in chain alarm control, but also lay the foundation for the later alarm and the activation of corresponding linkage fire extinguishing facilities. the

The future requirements for ship fire alarm systems are not only to realize the accuracy of cabin chain alarms, but also to make corresponding fire prevention measures and linkage fire extinguishing actions on this basis, so as to increase the accuracy of fire discrimination and control of the alarm system , effectively reducing the frequency of missed and false alarms, making the ship's fire alarm system more complete and efficient, and more intelligent and humane. the

Contents of the invention

The purpose of the present invention is to provide a method for obtaining accurate, timely and reliable interlocking alarm level information of cabins, advanced cabin fire information alarm, which can save precious time for later fire extinguishing, and avoid cabin related fires from causing serious damage to ships and personnel. Evaluation method for fire chain alarm level of lost ship cabins. The object of the present invention is also to provide a ship cabin fire chain alarm level evaluation device. the

Ship cabin fire chain alarm level evaluation method of the present invention comprises the following steps:

1) The centralized fire alarm controller collects the fire information obtained by the temperature and smoke detectors in the cabins adjacent to the fire cabin;

2) The centralized fire alarm controller reads the cabin fire hazard level information. The cabin fire hazard level information is based on the quantity and characteristics of combustibles in the ship’s cabin, the ventilation status of the cabin, and the state of the fire protection system. The factors are summarized and quantification, and based on the non-linear synthetic index of the generalized index system, it is used to measure the cabin fire risk;

3) Perform normalized preprocessing on the fire information obtained by temperature and smoke detectors and the cabin fire hazard level information;

4) Send the processed data to the fuzzy reasoning module for fuzzy reasoning, so as to obtain the evaluation result of cabin fire chain alarm level. the

The composition of the ship cabin fire chain alarm level evaluation device of the present invention includes a signal acquisition module of each cabin fire alarm element, a signal preprocessing module, a fuzzification module, a fuzzy reasoning judgment module and a defuzzification module; The fire information obtained by detectors and smoke detectors and the cabin fire hazard level information; the signal preprocessing module preprocesses the information and normalizes it to the range of [0,1]; the fuzzy reasoning module first Fuzzy with the output, establish the fuzzy set on the domain of discourse, design the membership function, transform the input and output, then establish the fuzzy control rule table, and finally generate the input and output table, according to the input, the cabin fire chain alarm level is obtained. Identify the fuzzy sets on the domain of discourse; the defuzzification module performs defuzzification to obtain the final result. the

The membership function is a triangular membership function. the

The form of the fuzzy rule is: IF (satisfying a set of conditions) THEN (pushing out a set of conclusions). the

In order to solve the defects of the ship fire alarm system in the chain alarm function of the cabin, the present invention proposes a fuzzy control-based evaluation method for the fire chain alarm level of the ship cabin. The reasoning system analyzes and processes the information of different types of detectors in each cabin and the fire hazard level information of the cabin itself, quickly and accurately gives the alarm level information of each cabin, and strives to achieve the accuracy, advanceness and comprehensiveness of the cabin chain alarm, Reduce major fire accidents caused by omission of cabin fire information. the

The test process of the present invention can utilize MATLAB and VC++ mixed programming to realize the combination of intelligent fusion algorithm and visual interface, and the performance of the system is experimentally verified, and the steps are as follows:

In the VC++ environment, a visual interface for cabin fire chain alarm level evaluation is designed. Four edit boxes are added to the interface, and three are used to input the preprocessed data information of temperature and smoke detector alarm signals and cabin fire hazard levels. The other is used to display the evaluation result information of the cabin fire chain alarm level. A fuzzy reasoning command button is added. When the button is clicked after inputting three alarm parameters, the cabin fire chain alarm level information will be displayed. the

The technical effect of the present invention is that the centralized fire alarm controller collects the alarm information of a certain cabin and simultaneously collects the fire information of its adjacent rooms, which embodies the chain alarm control idea of the fire alarm system, and utilizes the fire chain alarm level of the ship cabin based on fuzzy control The evaluation method has obtained reliable and accurate cabin alarm level information, expanded the range of associated alarms, improved the control accuracy of the system, realized the comprehensiveness, advancedness, accuracy and reliability of the ship fire alarm system alarms, and achieved the reduction of false alarms. , the purpose of the false alarm rate, effectively reducing the probability of chain fires in the cabins that may cause heavy losses to the ship. the

Description of drawings

Fig. 1 is a flow chart of cabin interlock alarm control of the ship fire automatic alarm system applying the ship cabin fire interlock alarm level evaluation method of the present invention. the

Fig. 2 is a structural diagram of a ship cabin fire chain alarm level evaluation system based on fuzzy control for realizing the ship cabin fire chain alarm level evaluation method of the present invention. the

Figure 3 is a triangular membership function curve. the

Detailed ways

Below in conjunction with accompanying drawing example the present invention will be further described:

The present invention is designed for the cabin chain alarm control of the ship automatic fire alarm system. The flow chart of the cabin chain alarm control of the ship fire automatic alarm system of the present invention is shown in Figure 1. The present invention is aimed at judging cabins 1 to 1 It is a ship cabin fire chain alarm level evaluation method designed to send the fire alarm level information of cabins 1 to N to the centralized alarm controller and determine whether to start the corresponding fire extinguishing facilities. the

In conjunction with Fig. 1, the chain alarm control flow chart of the ship fire automatic alarm system of the present invention shows the control idea of the ship fire alarm system chain alarm. The ship fire alarm system periodically collects the detector alarm signals of each cabin. When a certain cabin alarms, the alarm information of this cabin will be sent to the centralized alarm controller. At this time, in order to achieve the purpose of chain alarm control, the centralized alarm controller will immediately Collect the fire information obtained by the fire detectors in multiple cabins adjacent to the cabin, and judge the fire alarm level information of these cabins. According to different alarm level signals, the centralized alarm controller will give different alarm signals and start Corresponding linkage fire extinguishing facilities for fire extinguishing. In this way, the function of the ship's automatic fire alarm system is more perfect. The application of the chain alarm control method of the cabin can reduce the false alarm rate of the alarm system, improve the control accuracy of the system, and consider the fire related information of each cabin to be more comprehensive, so as to avoid accidents due to negligence. The huge loss caused to the whole ship due to the alarm information of each cabin. The ship cabin fire chain alarm level evaluation method designed by the present invention is applied to the ship cabin fire chain alarm control process to judge whether cabins 1 to N are alarming and send the cabin 1 to N fire alarm level information to the centralized alarm controller and determine Whether to activate the corresponding fire extinguishing facilities. the

In conjunction with FIG. 2 , it is a structure diagram of the cabin fire interlock alarm system based on fuzzy control of the present invention. It is mainly composed of signal acquisition, signal preprocessing, fuzzification, fuzzy reasoning and judgment, and defuzzification of each cabin fire alarm element, and finally obtains cabin fire chain alarm level information. the

The input signals of the system include the fire information obtained by the temperature detectors and smoke detectors of each cabin, and the cabin fire hazard level information obtained through domestic and foreign fire simulation software (the cabin fire hazard level can be found through relevant literature inquiries), The information is preprocessed and normalized to the range of [0,1], so that the next step of data processing can be performed. Collecting the above information can enable the system to identify the fire alarm and discrimination information in the cabin early, help prevent fire hazards, and reduce the rate of missed and false alarms. the

In order to improve the accuracy and anti-interference ability of cabin fire level discrimination, and make the output of the system closer to human's thinking and judgment, the present invention adopts fuzzy reasoning method to further process the preprocessed data. The fuzzy reasoning module first fuzzifies the input and output, establishes the fuzzy set on the domain of discourse, designs the corresponding membership function, and transforms the input and output accordingly. The triangular membership function is used here. Secondly, establish the fuzzy control rule table. Fuzzy control rules are the core of fuzzy control. These rules usually come from expert knowledge in related fields, and can also be given by a large number of experimental data. The fuzzy rule form of the present invention is: IF (satisfying a set of conditions) THEN (pushing out a set of conclusions). When formulating rules, it is necessary to ensure the completeness of fuzzy rules and the compatibility of fuzzy rules. Finally, the input and output table is generated, and the fuzzy set of cabin fire chain alarm level in the identification domain is obtained according to the input, and then defuzzification is performed to obtain the final result. the

The design idea of the fuzzy reasoning model is as follows:

1) Fuzzification of input and output

The input of the fuzzy controller is the fire information obtained by the temperature detectors and smoke detectors in each cabin and the cabin fire hazard level information collected by the centralized controller, and the output is the cabin fire chain alarm level information. After processing, it must first be converted into numerical information in the range of [0,1] as the domain of discourse. The normalization formula here is chosen as follows:

${\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}$

Among them, x∈[x _min ,x _max ], x _min and x _max are the maximum and minimum values of the input volume to be preprocessed respectively. Secondly, they should be fuzzified, that is, to determine the fuzzy membership functions of the three input variables respectively. However, the determination of the membership degree and membership function of fuzzy sets is complex and diverse, including expert determination methods, fuzzy statistics methods, and comparison and ranking methods. Since the experimental data of ship cabin fires is difficult to collect, there is no suitable "objective" scale, so Using expert determination. The upper and lower limits of the input and output are both set to [0,1]. According to the actual situation, experience and statistical analysis of fire data in the ship cabin, the fire information of the temperature detector, the fire information of the smoke detector, and the fire information of each cabin can be calculated. The fire risk degree and the output cabin fire chain alarm level are set as follows: high probability of fire occurrence (PB), medium probability of fire occurrence (PM), small probability of fire occurrence (PS) and no fire occurrence (ZO). The commonly used fuzzy membership functions are triangular membership function, normality membership function and trapezoidal membership function. Here, the membership function of fuzzy numbers is selected as triangular function, which is easy to calculate and widely used. The analytical formula of this function is:

The membership function curve is shown in Figure 3. the

2) Fuzzy logic reasoning

The present invention adopts Mamdani method to realize fuzzy logic reasoning, and the form of fuzzy rules is: IF (satisfying a set of conditions) THEN (pushing out a set of conclusions), specifically (there are 10 rules listed below):

1. IF (the fire information of the heat detector is ZO) AND (the fire information of the smoke detector is ZO) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is ZO)

2. IF (the fire information of the heat detector is PS) AND (the fire information of the smoke detector is ZO) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is PS)

3. IF (the fire information of the heat detector is PM) AND (the fire information of the smoke detector is ZO) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is PM)

4. IF (the fire information of the heat detector is PB) AND (the fire information of the smoke detector is ZO) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is PB)

5. IF (the fire information of the heat detector is ZO) AND (the fire information of the smoke detector is PS) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is ZO)

6. IF (the fire information of the heat detector is PS) AND (the fire information of the smoke detector is PS) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is PS)

7. IF (the fire information of the heat detector is PM) AND (the fire information of the smoke detector is PS) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is PM)

8. IF (the fire information of the heat detector is PB) AND (the fire information of the smoke detector is PS) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is PB)

9. IF (the fire information of the heat detector is ZO) AND (the fire information of the smoke detector is PM) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is ZO)

10. IF (the fire information of the heat detector is PS) AND (the fire information of the smoke detector is PM) AND (the fire hazard level of this cabin is ZO) THEN (the fire chain alarm level of this cabin is PS)

3) Defuzzification

The result obtained by fuzzy reasoning decision is a fuzzy set, but what the actual system needs to obtain is accurate cabin fire alarm level information, so defuzzification is required to obtain this accurate value. In fuzzy control, there are many methods of defuzzification, the commonly used ones are maximum membership function method, center of gravity method and median method, the present invention uses center of gravity method to defuzzify the output variable, thus obtaining accurate cabin Fire interlock alarm level output, the cabin fire interlock alarm level output can be set to: no alarm, pre-alarm and alarm. the

After designing the fuzzy controller, test multiple sets of data through the visual programming interface. By observing the experimental results, it can be seen that the fire chain alarm level evaluation method based on fuzzy control can make the centralized fire alarm controller know that a fire has occurred. At the same time, the cabin can accurately obtain the fire alarm level information of its adjacent cabins, reliably and comprehensively provide cabin fire chain alarm information for the staff on board, and indeed achieve early fire warning for each cabin, improving the control accuracy of the alarm system. the

Claims (2)

1. A ship cabin fire chain alarm level evaluation device is characterized in that: constitute a signal acquisition module comprising each cabin fire alarm element, a signal preprocessing module, a fuzzification module, a fuzzy reasoning judgment module and a defuzzification module; the signal acquisition module Collect the fire information obtained by the temperature detectors and smoke detectors in each cabin and the fire hazard level information of the cabin; the signal preprocessing module preprocesses the information and normalizes it to the range of [0,1]; fuzzy reasoning The module first fuzzifies the input and output, establishes the fuzzy set on the domain of discourse, designs the membership function, and transforms the input and output, then establishes the fuzzy control rule table, finally generates the input and output table, and obtains the cabin according to the input The fuzzy set of the fire chain alarm level in the identification domain; the defuzzification module performs defuzzification to obtain the final result. 2. The ship cabin fire chain alarm level evaluation device according to claim 1, characterized in that: said membership function is a triangular membership function.

Images