Heat exchange station safety management system and method based on artificial intelligence

Abstract

The invention discloses a heat exchange station safety management system and method based on artificial intelligence, and relates to the technical field of heat exchange station safety management, the system comprises an electric power safety management module, a control center and a one-key start control module; the electric power safety management module is used for monitoring a heat exchange station power loss signal and a power supply signal; the control center is used for controlling the heat exchange station; the one-key start control module is used for intelligently forming a one-key start control strategy and feeding back the one-key start control strategy to the control center after power loss and power supply recovery in the station; the output end of the electric power safety management module is connected with the control center and the one-key starting control module; and the control center is connected with the one-key starting control module. The safety problem caused by sudden power failure of the heat exchange station is optimized, the secondary water supply outlet electric blow-down valve is additionally arranged, the problem that the temperature is sharply increased after power failure can be effectively solved, meanwhile, a one-key starting strategy is provided, and the work recovery efficiency of the heat exchange station is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of heat exchange station safety management, and in particular to a heat exchange station safety management system and method based on artificial intelligence. Background Art

[0002] During the operation of the heating season, safety operation accidents caused by power outages often occur. Power outages will cause the secondary side circulation pump in the heat exchange station to stop running, causing the secondary water supply side temperature to rise. Tests have shown that the primary side water supply temperature can be reached in a relatively short time. If the circulation pump is turned on immediately after the heat exchange station is restored to power, high-temperature water will be directly delivered to the user, causing a safety accident of water leakage. In addition, if the circulation pump and water supply pump cannot be turned on as soon as possible after power is turned on, the room temperature of the heat user will drop, increasing the risk of complaints from heat users. Summary of the invention

[0003] The purpose of the present invention is to provide a heat exchange station safety management system and method based on artificial intelligence to solve the problems raised in the prior art.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a heat exchange station safety management system based on artificial intelligence, the system comprising a power safety management module, a control center and a one-button start control module;

[0005] The power safety management module is used to monitor the power failure signal and power supply signal of the heat exchange station; the control center is used to realize the control of the heat exchange station; the one-key start control module is used to intelligently form a one-key start control strategy and feedback it to the control center after the power failure and power restoration in the station; the control center includes a plate heat exchanger, a circulating water pump, a water supply pump, a softening water tank, a primary side three-dimensional expansion decontaminant and a secondary side three-dimensional expansion decontaminant;

[0006] The output end of the power safety management module is connected to the control center and the one-button start control module; the control center is connected to the one-button start control module.

[0007] According to the above technical solution, the power safety management module includes an uninterruptible power supply and a signal management unit;

[0008] The signal management unit is used to form a signal, send a power failure signal when the heat exchange station is out of power, and send a power supply signal when the power is restored; the uninterruptible power supply is used to temporarily maintain power supply when the heat exchange station is out of power.

[0009] According to the above technical solution, in the control center, one side of the plate heat exchanger is connected to the primary side high-temperature water return, and the other end of the same side is connected to the primary side three-dimensional expansion decontaminant; the other side is connected to the secondary side drainage, and the other end of the other side is connected to the circulating water pump; the other end of the circulating water pump is respectively connected to the make-up water pump and the secondary side three-dimensional expansion decontaminant; the other end of the make-up water pump is connected to the softening water tank; one side of the primary side three-dimensional expansion decontaminant is connected to the primary side high-temperature water inlet; one side of the secondary side three-dimensional expansion decontaminant is connected to the secondary side high-temperature water return.

[0010] According to the above technical solution, after the control center detects the power failure signal, it first closes the primary side three-dimensional expansion decontaminant and simultaneously opens the secondary side three-dimensional expansion decontaminant with a small flow rate. The value of the small flow rate is determined by the parameter value fed back by the one-button start control module.

[0011] According to the above technical solution, when the power supply in the station is restored, the one-button start control module intelligently forms a one-button start control strategy:

[0012] The one-key start control strategy includes a parameter interface;

[0013] The parameters include: opening time of electric drain valve, operating frequency of circulating water pump, constant pressure of water supply pump, opening of primary electric valve, low protection setting value of softening water tank level, high protection setting value of water supply temperature and low protection setting value of circulating water pump operating pressure;

[0014] The electric drain valve opens and drains water to lower the secondary water supply temperature according to the set opening time; it closes automatically after the set opening time is reached (if the set opening time is not reached, you can manually click the "Close" button on the drain valve to close it manually); when the control center detects that the softening water tank level is higher than the low protection setting value, the water supply pump is turned on;

[0015] When the control center detects that the secondary water supply temperature is lower than the high protection setting value and the secondary return water pressure is greater than the low protection setting value of the circulating water pump operating pressure, the circulating pump is turned on; otherwise, the circulating pump will not be turned on.

[0016] A heat exchange station safety management method based on artificial intelligence, the method comprising the following steps:

[0017] S1. The power safety management module of the heat exchange station monitors the power failure signal and power supply signal of the heat exchange station in real time to form the power outage duration;

[0018] S2. Retrieve the manual parameter setting values ​​of the administrator in the power outage restoration state in the historical big database to form a data group for each power outage restoration state, wherein the data group includes the power outage duration and the manual setting values ​​of each parameter;

[0019] S3. Build an analysis model for one - key startup control strategy, automatically set each parameter based on different power outage durations, and feedback to the administrator window for confirmation.

[0020] According to the above - mentioned technical solution, the one - key startup control strategy analysis model includes:

[0021] Obtain the data groups under each power - outage restoration power - supply state, sort the data groups in chronological order, and record each data group as a state. The initial state is recorded as S1, and the transition from one state to the next state is recorded as:

[0022] p(S t+1 丨S t ) = p(S t+1 丨A1, A2, A3, A4, A5, A6, A7, E0)

[0023] Among them, S t represents the t - th state; S t+1 represents the (t + 1) - th state; p represents the state - transition matrix; p(S t+1 丨S t ) represents that the next state S t+1 depends on the current state S t , and makes a transition according to the state - transition matrix; A1, A2, A3, A4, A5, A6, A7 refer to the parameters within the state, corresponding to the opening time of the electric sewage valve, the operating frequency of the circulating water pump, the constant - pressure water - supply pressure of the make - up water pump, the opening degree of the primary - side electric valve, the low - level protection setting value of the softening water tank, the high - temperature protection setting value of the water supply, and the low - pressure protection setting value of the circulating water pump operating pressure; E0 represents the power - outage duration;

[0024] Set the reward function R during the state - transition process. The value of the reward function is inversely proportional to the parameter difference. The parameter difference refers to the absolute value of the difference between the predicted value and the actual value formed according to the current state, and forms the return under reinforcement learning:

[0025]

[0026] Among them, M represents the total number of parameters; G m＝1，M represents the return value during the transition process; m represents the parameter serial number; γ represents the reward discount factor.

[0027] According to the above - mentioned technical solution, it also includes:

[0028] Build a value function V. The value function is directly proportional to the total return value. The total return value refers to the sum of the return values formed during the state - transition process of all data groups;

[0029] Since the state - transitions between different data groups are different, setting the transition from a certain state to a certain state as an action, different trajectories are formed based on different actions:

[0030] A certain action is denoted as c, and π is set [ t丨c ] is the probability of taking action c at state t;

[0031] Then the expected value of the returns of all possible trajectories under a given policy is expressed as:

[0032]

[0033] where J ( H ) represents the expected value; H represents the given policy; n represents the trajectory number; N represents the total number of trajectories; G ( H ) refers to the return under the given policy; L ( n丨H ) refers to the probability of the occurrence of trajectory n at one step in the Markov decision process given the initial state distribution x0 and policy H;

[0034]

[0035] where p ( S t+1 丨S t ,c ) refers to taking action c during the transition from the t-th state to the (t + 1)-th state; K refers to the step size; k refers to the serial number;

[0036] By using an optimization method to maximize the expected return, select the policy under the maximized expected return:

[0037] H0 = arg max J(H)

[0038] where H0 is the policy selected under the maximized expected return; after the policy is formed, based on the different actions therein, calculate the total sum of the formed return values. If the total sum of the return values is lower than the set threshold, feedback to the administrator port to remind to increase the data group; if the total sum of the return values is not lower than the set threshold, form the prediction parameters for the current power outage duration according to the policy and register them in the automatic setting, waiting for the administrator to review.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention optimizes the safety problems that will be caused by the sudden power outage of the heat exchange station, adds an electric sewage discharge valve at the secondary water supply outlet, can effectively improve the problem of sudden temperature increase after the power outage, reduce the safety accidents of water leakage caused by high-temperature water, and at the same time provides a one-key start strategy. By using the method of artificial intelligence and reinforcement learning, automatic setting of parameters is formed, which can greatly reduce the time and quickly restore the normal operation of the heat exchange station. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a connection diagram of an artificial intelligence-based heat exchange station safety management system of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Example: Figure 1 As shown, the present invention provides a technical solution, a heat exchange station safety management system based on artificial intelligence, the system includes a power safety management module, a control center and a one-button start control module;

[0043] The power safety management module is used to monitor the power failure signal and power supply signal of the heat exchange station; the control center is used to realize the control of the heat exchange station; the one-key start control module is used to intelligently form a one-key start control strategy and feedback it to the control center after the power failure and power restoration in the station; the control center includes a plate heat exchanger, a circulating water pump, a water supply pump, a softening water tank, a primary side three-dimensional expansion decontaminant and a secondary side three-dimensional expansion decontaminant;

[0044] The output end of the power safety management module is connected to the control center and the one-button start control module; the control center is connected to the one-button start control module.

[0045] The power safety management module includes an uninterruptible power supply and a signal management unit;

[0046] The signal management unit is used to form a signal, send a power failure signal when the heat exchange station is out of power, and send a power supply signal when the power is restored; the uninterruptible power supply is used to temporarily maintain power supply when the heat exchange station is out of power.

[0047] In the control center, one side of the plate heat exchanger is connected to the primary side high-temperature water return, and the other end of the same side is connected to the primary side three-dimensional expansion decontaminant; the other side is connected to the secondary side drainage, and the other end of the other side is connected to the circulating water pump; the other end of the circulating water pump is respectively connected to the make-up water pump and the secondary side three-dimensional expansion decontaminant; the other end of the make-up water pump is connected to the softening water tank; one side of the primary side three-dimensional expansion decontaminant is connected to the primary side high-temperature water inlet; one side of the secondary side three-dimensional expansion decontaminant is connected to the secondary side high-temperature water return.

[0048] After the control center detects the power failure signal, it first closes the primary side three-dimensional expansion decontaminant and simultaneously opens the secondary side three-dimensional expansion decontaminant with a small flow rate. The value of the small flow rate is determined by the parameter value fed back by the one-button start control module.

[0049] When the power supply in the station is restored, the one-key start control module intelligently forms a one-key start control strategy:

[0050] The one-key start control strategy includes a parameter interface;

[0051] The parameters include: the opening time of the electric sewage discharge valve, the operating frequency of the circulating water pump, the constant pressure water supply pressure of the make-up water pump, the opening degree of the primary side electric valve, the low protection setting value of the softening water tank liquid level, the high protection setting value of the water supply temperature, and the low protection setting value of the operating pressure of the circulating water pump;

[0052] Open the electric sewage discharge valve, and drain water according to the set opening time to reduce the secondary water supply temperature; after reaching the set opening time, it will automatically close (if the set opening time is not reached, you can manually click the "Close" button at the drain valve in the process flow diagram to close it manually); when the control center detects that the liquid level of the softening water tank is higher than the low protection setting value, turn on the make-up water pump;

[0053] When the control center detects that the secondary water supply temperature is lower than the high protection setting value and the secondary return water pressure is greater than the low protection setting value of the operating pressure of the circulating water pump, turn on the circulating pump; otherwise, the circulating pump will not be turned on.

[0054] Based on the system, a heat exchange station safety management method based on artificial intelligence is also provided. The method includes the following steps:

[0055] S1. The heat exchange station power safety management module monitors the power failure signal and power supply signal of the heat exchange station in real time to form the power outage duration;

[0056] S2. Retrieve the manual parameter setting values of the administrator under the power outage restoration power supply state in the historical big database to form a data group under each power outage restoration power supply state. The data group includes the power outage duration and the manual setting values of each parameter;

[0057] S3. Build a one-key start control strategy analysis model, and automatically set each parameter based on different power outage durations, and feedback to the administrator window for confirmation.

[0058] The one-key start control strategy analysis model includes:

[0059] Obtain the data group under each power outage restoration power supply state, sort the data group in chronological order, and record each data group as a state. The initial state is recorded as S1, and the transition from one state to the next state is recorded as:

[0060] p(S t+1 丨S t ) = p(S t+1 丨A1, A2, A3, A4, A5, A6, A7, E0)

[0061] Among them, St represents the t-th state; S t+1 represents the (t + 1)-th state; p represents the state transition matrix; p(S t+1 |S t ) represents the next state S t+1 depends on the current state S t , and makes a transition according to the state transition matrix; A1, A2, A3, A4, A5, A6, A7 refer to the parameters within the state, corresponding to the opening time of the electric sewage discharge valve, the operating frequency of the circulating water pump, the constant pressure makeup water pressure of the makeup water pump, the opening degree of the primary side electric valve, the low-level protection setting value of the softening water tank, the high-temperature protection setting value of the water supply, and the low-pressure protection setting value of the operating pressure of the circulating water pump; E0 represents the power outage duration;

[0062] Set the reward function R during the state transition process. The value of the reward function is inversely proportional to the parameter difference, where the parameter difference refers to the absolute value of the difference between the predicted value formed according to the current state and the actual value, forming the return under reinforcement learning:

[0063]

[0064] where, M represents the total number of parameters; G m＝1，M represents the return value during the transition process; m represents the parameter serial number; γ represents the reward discount factor.

[0065] It also includes:

[0066] Construct the value function V. The value function is directly proportional to the total return value, where the total return value refers to the sum of the return values formed during the state transition process of all data groups;

[0067] Since the state transitions between different data groups are different, setting the transition from a certain state to a certain state as an action, different trajectories are formed based on different actions:

[0068] A certain action is denoted as c, and set π [ t|c ] as the probability of taking action c at state t;

[0069] Then the expected value of the return of all possible trajectories under a given policy is expressed as:

[0070]

[0071] where, J ( H ) represents the expected value; H represents the given policy; n represents the trajectory serial number; N represents the total number of trajectories; G ( H ) refers to the return under the given policy; L ( n|H )Refers to the occurrence probability of trajectory n at one step in a Markov decision process given the initial state distribution x0 and policy H;

[0072]

[0073] where p ( S t+1 丨S t ,c ) refers to taking action c during the transition from the t-th state to the (t + 1)-th state; K refers to the step size; k refers to the serial number;

[0074] By using an optimization method to maximize the expected return, select the policy under the maximized expected return:

[0075] H0 = arg max J ( H )

[0076] where H0 is the selected policy under the maximized expected return; after the policy is formed, based on different actions therein, calculate the total return value formed. If the total return value is lower than the set threshold, feedback to the administrator port to remind to increase the data group; if the total return value is not lower than the set threshold, form the prediction parameters for the current power outage duration according to the policy and register them in the automatic setting, waiting for the administrator's review.

[0077] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. A heat exchange station safety management system based on artificial intelligence, characterized by: The system includes a power safety management module, a control center, and a one-button start control module; The power safety management module is used to monitor the power failure signal and power supply signal of the heat exchange station; the control center is used to realize the control of the heat exchange station; the one-key start control module is used to intelligently form a one-key start control strategy and feedback it to the control center after the power failure and power restoration in the station; the control center includes a plate heat exchanger, a circulating water pump, a water supply pump, a softening water tank, a primary side three-dimensional expansion decontaminant and a secondary side three-dimensional expansion decontaminant; The output end of the power safety management module is connected to the control center and the one-button start control module; the control center is connected to the one-button start control module.

2. The heat exchange station safety management system based on artificial intelligence according to claim 1 is characterized by: The power safety management module includes an uninterruptible power supply and a signal management unit; The signal management unit is used to form a signal, send a power failure signal when the heat exchange station is out of power, and send a power supply signal when the power is restored; the uninterruptible power supply is used to temporarily maintain power supply when the heat exchange station is out of power.

3. The heat exchange station safety management system based on artificial intelligence according to claim 1 is characterized by: In the control center, one side of the plate heat exchanger is connected to the primary side high-temperature water return, and the other end of the same side is connected to the primary side three-dimensional expansion decontaminant; the other side is connected to the secondary side drainage, and the other end of the other side is connected to the circulating water pump; the other end of the circulating water pump is respectively connected to the make-up water pump and the secondary side three-dimensional expansion decontaminant; the other end of the make-up water pump is connected to the softening water tank; one side of the primary side three-dimensional expansion decontaminant is connected to the primary side high-temperature water inlet; one side of the secondary side three-dimensional expansion decontaminant is connected to the secondary side high-temperature water return.

4. The heat exchange station safety management system based on artificial intelligence according to claim 1 is characterized by: After the control center detects the power failure signal, it first closes the primary side three-dimensional expansion decontaminant and simultaneously opens the secondary side three-dimensional expansion decontaminant with a small flow rate. The value of the small flow rate is determined by the parameter value fed back by the one-button start control module.

5. The heat exchange station safety management system based on artificial intelligence according to claim 1 is characterized by: When power is restored in the station, the one-button start control module intelligently forms a one-button start control strategy: The one-key start control strategy includes a parameter interface; The parameters include: opening time of electric drain valve, operating frequency of circulating water pump, constant pressure of water supply pump, opening of primary electric valve, low protection setting value of softening water tank level, high protection setting value of water supply temperature and low protection setting value of circulating water pump operating pressure; The electric drain valve opens and drains water to lower the secondary water supply temperature according to the set opening time; it closes automatically after the set opening time is reached; when the control center detects that the softening water tank level is higher than the low protection setting value, the water supply pump is turned on; When the control center detects that the secondary water supply temperature is lower than the high protection setting value and the secondary return water pressure is greater than the low protection setting value of the circulating water pump operating pressure, the circulating pump is turned on; otherwise, the circulating pump will not be turned on.

6. A heat exchange station safety management method based on artificial intelligence, using the heat exchange station safety management system based on artificial intelligence as claimed in claim 1, characterized in that: The method comprises the following steps: S1. The power safety management module of the heat exchange station monitors the power failure signal and power supply signal of the heat exchange station in real time to form the power outage duration; S2. Retrieving the manual parameter setting values ​​of the administrator in the power outage restoration state in the historical big database to form a data group for each power outage restoration state, wherein the data group includes the power outage duration and the manual setting values ​​of each parameter; S3. Build a one-button start control strategy analysis model to automatically set various parameters based on different power outage durations, and feed back to the administrator window for confirmation.

7. The heat exchange station safety management method based on artificial intelligence according to claim 6 is characterized by: The one-button start control strategy analysis model includes: Obtain each data group under the power restoration state of power outage, sort the data groups in chronological order, and record each data group as a state. The initial state is recorded as S1, and the transition from one state to the next state is recorded as: p ( S t+1 丨S t) =p ( S t+1 丨A1, A2, A3, A4, A5, A6, A7, E0) Among them, S t represents the t-th state; S t+1 represents the (t + 1)-th state; p represents the state transition matrix; p(S t+1 |S t ) represents that the next state S t+1 depends on the current state S t , and makes a transition according to the state transition matrix; A1, A2, A3, A4, A5, A6, A7 refer to the parameters within the state, corresponding to the opening time of the electric sewage discharge valve, the operating frequency of the circulating water pump, the constant pressure make-up water pressure of the make-up water pump, the opening degree of the primary side electric valve, the low-level protection setting value of the softening water tank, the high-temperature protection setting value of the water supply, and the low-pressure protection setting value of the operating pressure of the circulating water pump; E0 represents the power outage duration; Set the reward function R during the state transition process. The reward function value is inversely proportional to the parameter difference. The parameter difference refers to the absolute value of the difference between the predicted value formed according to the current state and the actual value, forming the return under reinforcement learning: Among them, M represents the total number of parameters; G m＝1，M represents the reward value during the transfer process; m represents the parameter number; γ represents the reward discount factor.

8. The heat exchange station safety management method based on artificial intelligence according to claim 7 is characterized in that: It also includes: Construct a value function V. The value function is directly proportional to the total return value. The total return value refers to the sum of the return values formed during the state transition process of all data groups; Since there are different state transitions between different data groups, setting the transition from a certain state to a certain state as an action, different trajectories are formed based on different actions: Record a certain action as c, and set π[t丨c] as the probability of taking action c at state t; Then the expected value of the return of all possible trajectories under a given policy is expressed as: Among them, J(H) represents the expected value; H represents the given policy; n represents the trajectory number; N represents the total number of trajectories; G(H) refers to the return under the given policy; L(n丨H) refers to the occurrence probability of trajectory n at a step in the Markov decision process given the initial state distribution x0 and policy H; where p(S t+1 | S t , c) means taking action c during the transition from the t-th state to the (t + 1)-th state; K refers to the step size; k refers to the serial number; Through an optimization method to maximize the expected return, select the policy under the maximized expected return: H0=arg maxJ ( H ) Among them, H0 is the selected policy under the maximized expected return; after the policy is formed, based on different actions among them, calculate the total return value formed. If the total return value is lower than the set threshold, feedback to the administrator port to remind to increase the data group; if the total return value is not lower than the set threshold, form the prediction parameters for the current power outage duration according to the policy, register them in the automatic setting, and wait for the administrator to review.

Images