Nuclear reactor risk guidance type multi-attribute autonomous operation decision-making method and system

Abstract

The invention provides a nuclear reactor risk guidance type multi-attribute autonomous operation decision-making method and system, and the method comprises the steps: building a plurality of alternative operation decisions of a nuclear reactor, the alternative operation decisions comprising expected loss of the nuclear reactor after control behaviors contained in the alternative operation decisions are executed, and the expected loss meets a preset loss condition; simulating the nuclear reactor on the basis of each alternative operation decision to obtain expected parameter values of operation state parameters of the nuclear reactor and expected utility values of corresponding effect variables in the simulation process, substituting the corresponding expected utility values into a utility function, and calculating to obtain actual utility values of all the utility variables; and determining a target operation decision of the nuclear reactor from a plurality of alternative operation decisions based on the actual utility value. According to the method, the influence of different alternative operation decisions on the nuclear reactor is quantified by adopting a multi-attribute utility theory algorithm, a rapid and accurate decision is made in a complex environment, and the labor cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear reactor operation decision-making, and in particular to a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method and system. Background Art

[0002] Nuclear reactor control systems are usually used to maintain the state parameters of nuclear reactors within the specified operating range. If these state parameters deviate from the specified range, it will cause transient conditions in nuclear power plants and pose a severe challenge to the accident mitigation system of nuclear power plants. Therefore, when transient conditions occur in nuclear reactors, how the reactor control system makes correct operating decisions to ensure the stable operation of nuclear reactors has become an important concern. However, the alternatives are diverse and numerous. In order to accurately determine the target decision, it takes a lot of effort to manually determine the operating decision, which has low decision-making efficiency and high labor costs.

[0003] At present, the nuclear reactor field usually makes static decisions through deterministic decision analysis. The deterministic decision analysis process does not consider the influence of randomness or uncertainty factors, and predicts the results of the decision based on certain parameters, which cannot capture the risks and uncertainties caused by parameter uncertainty or random fluctuations. Static decisions do not consider the dynamic changes of time, but analyze and select at a specific moment or state, resulting in the target operation decision derived from static decision making unable to adapt to the complex situation that changes over time, unable to consider possible changes in the future, and therefore not flexible enough and unable to respond to emergencies. Therefore, the accuracy and effectiveness of static decision making through deterministic decision analysis are poor. Summary of the invention

[0004] The technical problem to be solved by the present disclosure is to overcome the defects of the prior art, namely, the deterministic nuclear reactor operation decision-making method for a single target does not take into account uncertainty, cannot dynamically match the operation scenario and is not suitable for complex decision-making scenarios with multiple attributes, and to provide a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method and system.

[0005] The present invention solves the above technical problems through the following technical solutions:

[0006] In a first aspect, a nuclear reactor risk-guided multi-attribute autonomous operation decision method is provided, the nuclear reactor risk-guided multi-attribute autonomous operation decision method comprising:

[0007] Establishing a plurality of alternative operation decisions for a nuclear reactor, wherein the alternative operation decisions include expected losses of the nuclear reactor after executing control actions included in the alternative operation decisions, wherein the expected losses meet preset loss conditions;

[0008] The nuclear reactor is simulated based on each alternative operation decision to obtain expected parameter values ​​of the operation state parameters of the nuclear reactor and expected utility values ​​of the corresponding effect variables during the simulation process, and the corresponding expected utility values ​​are substituted into the utility function to calculate the actual utility values ​​of the utility variables; wherein the utility function represents the corresponding relationship between the operation state parameters and the utility variables, and the actual utility value of the utility variable corresponding to the state parameter whose parameter value does not exceed the safety range is greater than the actual utility value of the utility variable corresponding to the state parameter whose parameter value exceeds the safety range;

[0009] Based on the actual utility value, a target operating decision for the nuclear reactor is determined from a plurality of alternative operating decisions.

[0010] Optionally, the utility function is obtained by the following steps:

[0011] For each of the operating status parameters, a utility function is obtained by performing an affine transformation on the probability density distribution function of the operating status parameter.

[0012] Optionally, the probability density distribution function is represented by Gaussian distribution.

[0013] Optionally, the step of establishing a plurality of alternative operating decisions for the nuclear reactor comprises:

[0014] According to the success probability of each sequence of the decision tree, the alternative operation decisions are screened out; each sequence represents an operation decision; the success probability is related to the expected loss;

[0015] and / or, updating the success probability of the alternative operation decision in the decision tree according to the system equipment status of the nuclear reactor monitored in real time;

[0016] Arrange the sequences in reverse order according to the success probabilities, and determine the operation decisions corresponding to the sequences whose success probabilities are greater than a probability threshold as the alternative operation decisions; or determine the operation decisions corresponding to a preset number of sequences that are ranked first as the alternative operation decisions.

[0017] Optionally, the step of determining the target operating decision of the nuclear reactor from a plurality of alternative operating decisions based on the actual utility value comprises:

[0018] determining a weight corresponding to the utility variable;

[0019] For each of the alternative operation decisions, weighted summing the actual utility values ​​of the utility variables according to the weights, and calculating the composite utility of the alternative operation decisions based on the weighted results;

[0020] The alternative operating decision with the maximum composite utility is determined as the target operating decision.

[0021] Optionally, the step of calculating the composite utility of the alternative operation decision based on the weighted result includes:

[0022] The success probability is multiplied by the weighted result to obtain the composite utility of the alternative operation decision.

[0023] Optionally, the effect variable is obtained by the following formula:

[0024]

[0025] Among them, x i is the i-th utility variable, p i is the i-th operating state parameter; (p i ) max is the maximum value of the i-th operating state parameter, (p i ) min is the minimum value of the i-th operating state parameter;

[0026] And / or, the weight is obtained by the following formula:

[0027]

[0028] Where N represents the total number of utility variables, ω j represents the weight corresponding to the j-th utility variable;

[0029] And / or, the expression of the probability density distribution function is as follows:

[0030]

[0031] Wherein, p(x|μ,σ) represents the probability density distribution function, x represents the utility variable, μ represents the mean value of the probability density distribution, σ represents the standard deviation of the probability density distribution, and e represents the base of the natural logarithm;

[0032] And / or, the utility function is expressed as follows:

[0033]

[0034] Wherein, u(x|μ,σ) represents the utility function, a and b are transformation coefficients of affine transformation, x represents the utility variable, μ represents the mean value of probability density distribution, σ represents the standard deviation of probability density distribution, and e represents the base of natural logarithm; the transformation coefficients a and b of affine transformation are calculated by the following formula:

[0035] a+b=1

[0036]

[0037] And / or, the composite utility is obtained by the following formula:

[0038]

[0039] Among them, U represents the composite utility, N represents the total number of the utility variables, ωi represents the weight corresponding to each utility variable, xi represents the i-th utility variable, and u(xi) represents the value of the dependent variable of the utility function, that is, the actual utility value of the utility variable.

[0040] In a second aspect, a nuclear reactor risk-guided multi-attribute autonomous operation decision system is provided. The nuclear reactor operation decision analysis system based on the multi-attribute utility theory includes:

[0041] A probabilistic decision analysis module, used to establish a plurality of alternative operation decisions for a nuclear reactor, wherein the alternative operation decisions include an expected loss of the nuclear reactor after executing the control behavior included in the alternative operation decision, and the expected loss satisfies a preset loss condition;

[0042] A deterministic decision analysis module is used to simulate the nuclear reactor based on each alternative operation decision to obtain the expected parameter value of the operation state parameter of the nuclear reactor and the expected utility value of the corresponding effect variable during the simulation process, and substitute the corresponding expected utility value into the utility function to calculate the actual utility value of each utility variable; wherein the utility function represents the corresponding relationship between the operation state parameter and the utility variable, and the actual utility value of the utility variable corresponding to the state parameter whose parameter value does not exceed the safety range is greater than the actual utility value of the utility variable corresponding to the state parameter whose parameter value exceeds the safety range;

[0043] The optimal decision generating module is used to determine the target operation decision of the nuclear reactor from a plurality of alternative operation decisions based on the actual utility value.

[0044] Optionally, the utility function is obtained by the following steps: for each of the operating state parameters, the utility function is obtained by performing an affine transformation on the probability density distribution function of the operating state parameter.

[0045] Optionally, the probability density distribution function is represented by Gaussian distribution.

[0046] Optionally, the probabilistic decision analysis module includes:

[0047] A decision screening unit, configured to screen out the candidate operation decisions according to the success probability of each sequence of the decision tree; each sequence represents an operation decision; the success probability is related to the expected loss;

[0048] and / or, a success probability updating unit, configured to update the success probability of the alternative operation decision in the decision tree according to the system equipment status of the nuclear reactor monitored in real time;

[0049] And / or, a sorting unit is used to arrange the sequence in reverse order according to the success probability, and determine the operation decisions corresponding to the sequences whose success probability is greater than a probability threshold as the alternative operation decisions; or, the sorting unit is used to determine the operation decisions corresponding to a preset number of sequences that are ranked at the top as the alternative operation decisions.

[0050] Optionally, the optimal decision generation module includes:

[0051] A weight determination unit, used to determine the weight corresponding to the utility variable;

[0052] a weighted summing unit, configured to perform weighted summing of the actual utility values ​​of the utility variables according to the weights for each of the alternative operation decisions, and calculate the composite utility of the alternative operation decisions based on the weighted results;

[0053] The target operation decision determining unit is used to determine the alternative operation decision with the maximum composite utility as the target operation decision.

[0054] Optionally, the weighted summation unit is specifically used to: multiply the success probability by the weighted result to obtain the composite utility of the alternative operation decision.

[0055] According to a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein when the processor executes the computer program, the nuclear reactor risk-guided multi-attribute autonomous operation decision-making method described in any one of the above items is implemented.

[0056] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the nuclear reactor risk-guided multi-attribute autonomous operation decision-making method described in any one of the above items is implemented.

[0057] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.

[0058] The positive progressive effect of the present disclosure lies in: the present disclosure adopts a multi-attribute utility theory algorithm as a mathematical analysis tool, which can comprehensively consider multiple operating status parameters, measure the value of multiple alternative operating decisions by establishing a unified scale, or measure the decision maker's satisfaction with multiple alternative operating decisions, realize cross-standard comparison and decision-making, calculate the actual utility value of the utility variable, quantify the impact of different alternative operating decisions on the nuclear reactor through the actual utility value, obtain accurate target operating decisions, and make fast and accurate decisions in a complex and changing environment to ensure the safe and stable operation of the nuclear reactor, improve the safety and safety management level of the nuclear reactor, greatly reduce labor costs, improve the scientificity and accuracy of decision-making, reduce decision-making risks, and provide strong support for engineering practice.

[0059] Furthermore, the present invention can update the success probability of alternative operating decisions in the decision tree according to the system equipment status of the nuclear reactor monitored in real time, make dynamic decisions, and take into account possible changes in the future, so that the determined target operating decisions can adapt to complex situations that change over time.

[0060] Furthermore, based on the multi-attribute utility theory, the present disclosure realizes multi-objective autonomous operation decision-making for nuclear reactors. By integrating multiple objectives, more comprehensive and balanced target operation decisions can be found, thereby improving the quality and effectiveness of decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flowchart of a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0062] Figure 2 A success probability graph of multiple sequences of a decision tree in a nuclear reactor risk-guided multi-attribute autonomous operation decision method provided by an exemplary embodiment of the present disclosure;

[0063] Figure 3 A relationship diagram between utility variables and operating state parameters in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0064] Figure 4 Another flow chart of a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0065] Figure 5 A flowchart of weight optimization in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0066] Figure 6A distribution curve diagram of a probability density distribution function in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0067] Figure 7 A schematic diagram of a utility function after affine transformation in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0068] Figure 8 A schematic diagram of a calculation process in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0069] Fig. 9 A framework diagram of a nuclear reactor risk-guided multi-attribute autonomous operation decision method provided by an exemplary embodiment of the present disclosure;

[0070] Fig.10 A schematic diagram of the relationship between core outlet temperature and time in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0071] Fig.11 A schematic diagram of the relationship between the utility variable corresponding to the core outlet temperature and time in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0072] Fig.12 A schematic diagram of utility functions corresponding to various utility variables in a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided by an exemplary embodiment of the present disclosure;

[0073] Fig.13 A schematic diagram of a module of a nuclear reactor risk-guided multi-attribute autonomous operation decision system provided by an exemplary embodiment of the present disclosure;

[0074] Fig.14 The present invention provides a schematic structural diagram of an electronic device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0075] The present disclosure is further described below by way of examples, but the present disclosure is not limited to the scope of the examples.

[0076] Prefixes such as "first" and "second" are used in the embodiments of the present disclosure only to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of prefixes such as ordinal numbers to distinguish description objects in the embodiments of the present disclosure does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and no unnecessary limitation should be constituted due to the use of such prefixes. In addition, in the description of the present embodiment, unless otherwise specified, the meaning of "plurality" is two or more.

[0077] Example 1

[0078] Decision making is one of the basic cognitive processes of human behavior. It is the process of determining and selecting alternatives. In this process, people can select a target decision from multiple alternatives according to specific criteria to control the target object. Each alternative provides a different method or path. By executing the alternative, the target object can be transformed from a preset state or condition to an ideal state or condition.

[0079] Among them, the target object can be selected according to the actual situation. The alternative scheme in this embodiment is specifically an alternative operation decision, which can be selected according to the actual situation and is not particularly limited here. The target object in this embodiment is a nuclear reactor. It should be noted that the nuclear reactor risk-guided multi-attribute autonomous operation decision method in this embodiment can not only be used to control nuclear reactors, but also can be used to control other target objects, which can be selected according to the actual situation and is not particularly limited here.

[0080] In engineering practice, the determination of target decisions is an important part of solving decision-making problems. Due to the nature of some decision-making problems, the alternatives are diverse and numerous, and a large number of alternatives need to be considered comprehensively. In order to accurately determine the target decision, a lot of manpower costs are required, and manual judgment is subjective, which cannot ensure the accuracy of the target decision. Especially in the face of complex decision-making scenarios with multiple objectives and multiple alternatives, how to comprehensively consider multiple operating status parameters and make comprehensive and scientific decisions scientifically and effectively, and determine accurate target decisions, has become an important issue.

[0081] Nuclear reactor control systems are generally designed to maintain the operating parameters of nuclear reactors within a specified operating range. The operating parameters include at least one of the following: reactor power, coolant flow, power-to-flow ratio, reactor outlet temperature, coolant level, and turbine status. The operating parameters listed above are only examples and can be selected according to actual conditions, and are not particularly limited here.

[0082] If the parameter value of the operating state parameter exceeds the specified safety range, for example, the parameter value is greater than the upper limit of the safety range or the parameter value is less than the lower limit of the safety range, it will cause a transient situation in the nuclear power plant and pose a severe challenge to the accident mitigation system of the nuclear power plant, including the challenge to the safety system of the nuclear power plant. Therefore, when a nuclear reactor experiences a transient situation, how the nuclear reactor control system makes correct decisions and executes the decision-making operation plan to ensure the stable operation of the reactor has become an important concern.

[0083] Utility theory is mainly used in economics to describe an individual's preference or satisfaction with a certain choice or result. Multi-attribute utility theory is an efficient quantitative comparison method and an indispensable and important part of modern decision science. With attributes as evaluation criteria, multi-attribute utility theory is suitable for evaluating complex decision scenarios with multiple attributes and a large number of alternative operational decisions. Multi-attribute utility theory quantifies each evaluation indicator into a utility value, multiplies the utility value of each alternative operational decision by the corresponding weight, and then performs weighted synthesis based on the synthesis model to calculate the composite utility. Following the principle of utility maximization, the alternative operational decision with the largest composite utility can be selected as the target operational decision.

[0084] Multi-attribute utility theory is usually applicable to decision-making problems with multiple attributes or multiple standards for alternatives, and is an important tool for decision analysis in the engineering field. It realizes cross-standard comparison and decision-making by converting quantitative data under different standards into common utility values. When dealing with complex decision-making problems, multi-attribute utility theory shows its unique advantages in many aspects. In the decision-making process, it realizes cross-standard comparison through dimensionless transformation, constructs utility functions to independently calculate the utility values ​​of each attribute, and considers the risk attitude and preference of the decision maker through weighting.

[0085] When a nuclear reactor encounters a transient situation, through the multi-attribute utility theory, multiple alternative operating decisions can be introduced into the decision analysis for evaluation with very little effort and without increasing too much workload. Once the utility function is constructed, the actual utility value can be obtained based on the utility function, and the alternative operating decisions can be evaluated based on the actual utility value. When there are a large number of alternative operating decisions, any number of alternative operating decisions can be evaluated without quantity restrictions, and then the target operating decision can be determined, which greatly improves decision-making efficiency.

[0086] The present invention proposes a nuclear reactor risk-guided multi-attribute autonomous operation decision-making method, which adopts multi-attribute utility theory as a mathematical analysis tool to ensure that multiple attributes and factors can be comprehensively considered, thereby constructing a structured comprehensive decision-making analysis framework.

[0087] Figure 1A flowchart of a nuclear reactor risk-guided multi-attribute autonomous operation decision method provided by an exemplary embodiment of the present disclosure, the nuclear reactor risk-guided multi-attribute autonomous operation decision method comprising:

[0088] Step S101, establishing a plurality of alternative operation decisions for a nuclear reactor, wherein the alternative operation decisions include expected losses of the nuclear reactor after executing control actions included in the alternative operation decisions, wherein the expected losses meet preset loss conditions;

[0089] Among them, the operating state parameter is used to detect whether the nuclear reactor fails, and the expected utility value is within the preset range. Each alternative operation decision is mainly composed of a varying number of control behaviors, including the component information of the nuclear reactor failure or performance degradation before the execution of the alternative operation decision, as well as the sensor and reactor status information. After the execution, the nuclear reactor can resume steady-state operation. For example, assuming that the control behavior includes reducing the power of the nuclear reactor, before the execution of the alternative operation decision, the sensor detects that the nuclear reactor fails or the performance degrades, and after the execution of the alternative operation decision, the nuclear reactor can resume steady-state operation.

[0090] The probabilistic decision analysis module is used to execute step 101, and the alternative operation decision obtained is sent to the deterministic decision analysis module for executing step 102. Although the probabilistic decision analysis module automatically generates a corresponding set of control behaviors for each alternative operation decision, it cannot determine the performance parameters of the components associated with the control behaviors. The control behaviors generated by the probabilistic decision analysis module only contain abstract action concepts without specific instructions. For example, the control behavior may be to reduce power, but it does not specify how much power needs to be reduced. Therefore, the alternative operation decisions determined by the probabilistic decision analysis module need to be input into the deterministic decision analysis module for further evaluation and analysis.

[0091] The expected loss meeting the preset loss condition may be that the expected loss is minimized, or the expected loss is less than a loss threshold. The loss threshold may be selected according to actual conditions and is not particularly limited here.

[0092] See also Figure 2 The decision tree in the present invention is shaped like an event tree in probabilistic safety assessment. The difference from the event tree in probabilistic safety assessment is that the expected consequence of the decision tree in the present invention is to avoid triggering a shutdown and calculate the success probability of each sequence rather than the frequency of core damage.

[0093] Figure 2The steam turbine control valve moves in the closing direction, the steam turbine control valve is reset, SGBV1 is opened, SGBV2 is opened, the feed water flow of the turbine bypass valve, the feed water flow to the steam generator 1 is reduced, the feed water flow to the steam generator 2 is reduced, the steam turbine bypass valve 1 is opened, the steam turbine bypass valve 2 is opened, the power of the reactor module 1 is reduced, the power of the reactor module 2 is reduced, the steam turbine bypass valve 1 is closed, the steam turbine bypass valve 2 is closed, the reactor module 1 is manually shut down, and the reactor module 2 is manually shut down constitute alternative operation decisions, that is, each sequence of the decision tree.

[0094] It should be noted that the above sequence is only used as an example and can be selected according to actual conditions and is not particularly limited here.

[0095] The state evolution of the nuclear reactor can be monitored by operating state parameters. The number and type of operating state parameters can be selected according to actual conditions and are not particularly limited here.

[0096] Table 1 lists some operating parameters of a liquid metal reactor and their rated steady-state values.

[0097] Table 1 Some system design variables and corresponding rated steady-state values ​​of an advanced liquid metal reactor

[0098]

[0099] Considering the complexity and potential safety risks of nuclear reactor systems, an effective decision-making method must be able to comprehensively consider operating status parameters of multiple different dimensions. The selection of operating status parameters needs to consider multiple factors, such as the operating status of the nuclear reactor, the performance of the cooling system, the temperature distribution of the fuel rods, etc. The specific selection can be based on actual conditions and is not specifically limited here.

[0100] Since the dimensions of multiple operating state parameters are different, it is difficult to measure them with a single indicator. Therefore, analyzing the operation decision of nuclear reactors based on multi-attribute utility theory has potential practical value.

[0101] The selection of operating status parameters generally selects shutdown variables that are within a safe range. Shutdown variables are variables that can be used to determine whether a nuclear reactor has shut down. The basic goal of the nuclear reactor control system in selecting target operation decisions is to keep the state of the nuclear power plant within a controllable state. If the challenge surface surrounded by the shutdown variables is exceeded, the nuclear reactor protection system and / or the dedicated safety facility drive system will be activated. Table 2 lists the shutdown variables and their functions of a liquid metal reactor. Failures include but are not limited to shutdowns, which can be selected according to actual conditions and are not specifically limited here. A controllable state means that no failure has occurred in the nuclear power plant. The monitoring variables in Table 2 are shutdown variables, and the safety function is which type of monitoring object the monitoring variable specifically monitors.

[0102] Table 2 Shutdown variables and related safety functions of an advanced liquid metal reactor

[0103]

[0104] The basic goal of the reactor control system is to keep the state of the nuclear power plant within the controllable range. If the challenge surface surrounded by the shutdown variables is exceeded, the reactor protection system and / or the dedicated safety facility driving system will be activated.

[0105] Since there are many types of operating state parameters and they have different measurement units (dimensions), it is difficult to directly judge the pros and cons of each alternative operating decision by numerical value, which affects the reliability of the target operating decision. When conducting decision analysis, it is necessary to eliminate the non-additivity caused by the different measurement units of each operating state parameter, convert the statistical indicators into additive values, and establish the corresponding relationship between the operating state parameters and the utility variables in the alternative operating decision.

[0106] Figure 3 is the relationship between a certain operating state parameter and the utility variable, where the horizontal axis is the parameter value of the operating state parameter and the vertical axis is the expected utility value of the utility variable.

[0107] In one embodiment, in order to more intuitively compare the effects of different alternative operating decisions on the operating state parameters, the operating state parameters need to be normalized, that is, the operating state parameters are transformed to the interval [0, 1].

[0108] Using the standard 0-1 transformation, the operating state parameters are linearly transformed to obtain the utility variables using the following formula:

[0109]

[0110] Among them, xi is the i-th utility variable, pi is the i-th operating state parameter; (pi)max is the maximum value of the i-th operating state parameter, and (pi)min is the minimum value of the i-th operating state parameter. The safety range can be determined according to the technical specifications of the nuclear power plant, and can be set according to the actual situation, and is not specifically limited here. Table 3 lists the utility variables of a liquid metal reactor.

[0111] Table 3 Utility variables of an advanced liquid metal reactor

[0112]

[0113] The safety variables in Table 3 are the operating state parameters. The maximum value refers to the maximum value of the operating state parameter's variation range, the minimum value refers to the minimum value of the operating state parameter's variation range, and the upper boundary and the lower boundary are the values ​​of the horizontal axis when the vertical axis of the probability density distribution function of the Gaussian distribution is 0.

[0114] Step S102: Simulate the nuclear reactor based on each alternative operation decision to obtain the expected parameter values ​​of the operating state parameters of the nuclear reactor and the expected utility values ​​of the corresponding effect variables during the simulation process, and substitute the corresponding expected utility values ​​into the utility function to calculate the actual utility values ​​of each utility variable.

[0115] Among them, the utility function represents the corresponding relationship between the operating state parameters and the utility variables, and the actual utility value of the utility variable corresponding to the operating state parameters whose parameter values ​​do not exceed the safety range is greater than the actual utility value of the utility variable corresponding to the operating state parameters whose parameter values ​​exceed the safety range.

[0116] The utility function needs to be able to reflect the impact of different values ​​of the operating state parameters on the quality of alternative operating decisions. When applying utility theory in the economic field, the range of expected utility values ​​is [0,1]. However, when applying utility theory in the field of nuclear reactors, one of the decision-making goals is to ensure the safety and stability of nuclear reactors, and constraints need to be met in the decision-making process. If the operating state parameters exceed the preset safety range, it is necessary to use negative actual utility values ​​to provide risk warnings for alternative operating decisions. Therefore, when applying utility theory in the field of nuclear reactors, the nuclear reactor risk-guided multi-attribute autonomous operation decision-making method in this embodiment expands the range of actual utility values ​​to [-∞,1].

[0117] The constraints and the range of actual utility values ​​are further explained below.

[0118] The constraint condition is that the operating state parameters cannot exceed the preset safety range. The safety range, the lower limit value and the upper limit value of the safety range can be selected according to the actual situation and are not particularly limited here. It can be understood that if the operating state parameters exceed the safety range, it means that the stable operation of the nuclear reactor cannot be guaranteed and the nuclear reactor may fail. Therefore, it is necessary to ensure that after the nuclear reactor is controlled according to various alternative operation decisions, the operating state parameters are within the safety range.

[0119] Therefore, when constructing the utility function, for the operating state parameters whose parameter values ​​are at the boundary of the safety range, that is, the parameter values ​​of the operating state parameters are at the lower and upper limits of the safety range, the actual utility value is 0; for the operating state parameters whose parameter values ​​are within the safety range, a higher actual utility value should be given, and for the operating state parameters whose parameter values ​​exceed the safety range, a lower actual utility value should be given. Once the parameter value of the operating state parameter exceeds the safety range, the actual utility value is reduced to a negative value, or even negative infinity, to avoid the decision-making process from selecting alternative operating decisions that may lead to safety risks. When applying utility theory in the field of nuclear reactors, the range of actual utility values ​​is extended to [-∞,1].

[0120] When constructing the utility function, within the safety range defined by the lower limit and the upper limit, the value of the dependent variable of the utility function needs to be positive, and in the area beyond the safety range, the value of the dependent variable of the utility function is negative. It can be understood that if the operating state parameter exceeds the safety range, the actual utility value of the utility variable corresponding to the operating state parameter is small.

[0121] In the multi-attribute utility theory, a utility function is constructed for each operating status parameter. The parameter value of the operating status parameter is converted into the actual utility value of the utility variable through the utility function. That is, the quantitative data under different standards are uniformly converted into a dimensionless value between 0 and 1, which plays a normalization role and reflects the decision maker's preference or satisfaction with different operating status parameters of different alternative operating decisions, thereby realizing cross-standard comparison and decision-making.

[0122] Step S103: Determine a target operation decision for the nuclear reactor from a plurality of alternative operation decisions based on the actual utility value.

[0123] See also Figure 4 The above-mentioned nuclear reactor risk-guided multi-attribute autonomous operation decision method can be implemented through a decision system, which is integrated into the nuclear reactor control system. The decision system is specifically used for the two functions of real-time monitoring and automatic start of the decision process. The decision system monitors the operating status of the nuclear reactor in real time, that is, monitors the various operating status parameters of the nuclear reactor, and determines whether the nuclear reactor fails according to the operating status parameters. When the nuclear reactor fails, the decision process is automatically started, that is, the above-mentioned nuclear reactor risk-guided multi-attribute autonomous operation decision method is executed, and the target operation decision of the nuclear reactor is determined from multiple alternative operation decisions. It can quickly identify problems and take corresponding control measures, effectively prevent accidents, and ensure the safe operation of nuclear reactors. It can not only improve the accuracy and efficiency of reactor fault control, but also reduce decision errors caused by human factors, and better deal with various uncertainties and emergencies.

[0124] Among them, the definition of the fault state can be set according to the actual situation. For example, the parameter values ​​of a preset number of operating status parameters exceed the state threshold. The preset number and state threshold can be set according to the actual situation and are not particularly limited here.

[0125] In this embodiment, a multi-attribute utility theory algorithm is used as a mathematical analysis tool, which can comprehensively consider multiple operating status parameters, measure the value of multiple alternative operating decisions by establishing a unified scale, or measure the decision maker's satisfaction with multiple alternative operating decisions, to achieve cross-standard comparison and decision-making, quantify the impact of different alternative operating decisions on the nuclear reactor through actual utility values, determine the target operating decision based on the actual utility value, and thus obtain an accurate target operating decision. It is possible to make fast and accurate decisions in a complex and changing environment to ensure the safe and stable operation of the nuclear reactor, improve the safety and safety management level of the nuclear reactor, greatly reduce labor costs, improve the scientificity and accuracy of decisions, reduce decision risks, provide strong support for engineering practice, and integrate the current operating status parameters and expected operating status parameters of the nuclear reactor into the process of formulating the target operating decision to generate a unique target operating decision.

[0126] Multi-attribute utility theory, as a decision analysis framework, allows decision makers to evaluate the utility of different options in complex decision problems with multiple attributes or characteristics, comprehensively considering the impact of multiple dimensions in order to make the best choice.

[0127] During the operation of nuclear reactors, there are many potential risk challenges, such as equipment aging, operational errors, changes in the external environment, and other multiple risk challenges. The multi-attribute utility theory can help managers quantify risks and assign weights, so as to accurately determine target operation decisions. Through the multi-attribute utility theory, multiple aspects such as the safety, economy, and environmental impact of nuclear reactors can be taken into consideration to achieve a comprehensive evaluation.

[0128] In today's complex engineering systems, the decision-making process often involves multiple interrelated attributes, that is, operating state parameters, and needs to comprehensively consider multiple aspects such as risk, performance, and cost. In the field of nuclear power, especially when a nuclear reactor experiences transient conditions, the success or failure of the target operation decision is not only related to the stability and safety of the nuclear reactor, but also directly related to the safety of the surrounding personnel and the stability of the environment. The risk-guided multi-attribute autonomous operation decision-making method of the risk-guided nuclear reactor in this embodiment can achieve scientific and effective decision-making in a complex nuclear reactor environment.

[0129] In one embodiment, step S103 includes: determining a weight corresponding to the utility variable.

[0130] Among them, the weight corresponding to each utility variable is equal.

[0131] The uniform weight distribution method is adopted to give equal weight to each utility variable, that is, to give equal importance to each utility variable. The weight can be calculated by the following formula:

[0132]

[0133] Where N represents the total number of utility variables, ω j Represents the weight corresponding to the j-th utility variable.

[0134] In this embodiment, a uniform weight distribution method is adopted to assign equal weight to each utility variable, and to attach equal importance to each utility variable, so as to ensure that each utility variable is treated equally in the decision-making process, and to avoid decision-making deviations caused by subjective bias or improper weight distribution to a certain extent, so as to ensure the safe and stable operation of the nuclear reactor and improve the safety and safety management level of the nuclear reactor.

[0135] In order to more comprehensively reflect the actual contribution of each utility variable in the decision-making process and improve the scientificity and accuracy of the decision, other more sophisticated and effective methods can be introduced to optimize the assignment of weights to ensure that each utility variable is assigned an appropriate weight, so that the determined target operation decision is more in line with the actual situation and meets the needs of the project.

[0136] like Figure 5 As shown, Figure 5 The attribute information in is the operating state parameter, the fault scenario represents the scenario where the nuclear reactor fails, and the optimal decision result and the historical optimal decision both represent the target operating decision obtained in history, which can be used as a reference for determining the current target operating decision. Under the premise of a large amount of applicable historical data (such as the optimal decision results and attribute information under the fault scenario) and expert knowledge, the following weight assignment optimization ideas can be proposed: define the utility function and derive the composite utility to reflect the performance of the decision under different weight combinations, that is, to reflect the composite utility of the alternative operating decisions under different weight combinations. Based on historical data and expert knowledge, the optimal weight combination is iterated through the target optimization algorithm. During the verification process, the weight combination consistent with the historical optimal decision is retained, and the fault scenario and the corresponding weight combination are output.

[0137] It can be understood that when determining the historical optimal decision, a process of weight determination will be experienced. The historical optimal decision is obtained based on the actual utility value and weight calculation. The weight combination consistent with the historical optimal decision is retained, that is, the weight combination that is the same as the weight combination used to determine the historical optimal decision is retained.

[0138] If the same failure occurs in a nuclear reactor later, the weight combination can be applied to the same failure scenario to achieve accurate decision-making. The entire process combines data-driven and expert knowledge, aiming to improve the scientificity and accuracy of decision-making by optimizing weight distribution.

[0139] The target optimization algorithm can be selected according to the actual situation and is not particularly limited here.

[0140] In one embodiment, the utility function is obtained by the following steps: for each operating state parameter, the utility function is obtained by performing an affine transformation on the probability density distribution function of the operating state parameter.

[0141] In this embodiment, according to the specific characteristics of the nuclear reactor field, a utility function is constructed through affine transformation, which realizes the quantification of the utility value of the utility variable, and then a comprehensive decision analysis can be performed based on the actual utility value, thereby improving the scientificity and accuracy of the decision, reducing the decision risk, and providing strong support for engineering practice.

[0142] In one embodiment, the probability density distribution function is represented by Gaussian distribution.

[0143] In order to make the utility theory be effectively applied in the engineering field, the definition of the utility function in the utility theory is extended to meet specific needs and requirements.

[0144] In practical applications in the field of engineering, since the parameter values ​​of the operating state parameters of nuclear reactors present the characteristics of random distribution, and Gaussian distribution, as a continuous probability distribution, can better fit this randomness, Gaussian distribution is used to represent the probability density distribution function of the operating state parameters of nuclear reactors.

[0145] The probability density distribution function is expressed as follows:

[0146]

[0147] Among them, p(x|μ,σ) represents the probability density distribution function, x represents the utility variable, μ represents the mean value of the probability density distribution, σ represents the standard deviation of the probability density distribution, and e represents the base of the natural logarithm.

[0148] Figure 6 A normal distribution graph with different standard deviations is shown, where the horizontal axis is the utility variable x, and the vertical axis is the probability density distribution function p(x|μ,σ), μ=0.5.

[0149] In this embodiment, μ=0.5, which can be selected according to actual conditions and is not particularly limited here.

[0150] The probability density distribution function of the Gaussian distribution is transformed by the following formula:

[0151]

[0152] Among them, u(x|μ,σ) represents the utility function, a and b are the transformation coefficients of the affine transformation, x represents the utility variable, μ represents the mean value of the probability density distribution, σ represents the standard deviation of the probability density distribution, and e represents the base of the natural logarithm.

[0153] Figure 7 The utility function obtained after affine transformation is shown, the vertical axis represents the utility function, the horizontal axis represents the utility variable, and the curve of the utility function intersects the horizontal axis at two points, which are the lower limit value and the upper limit value of the safety state respectively.

[0154] The transformation coefficients a and b of the affine transformation can be obtained by solving the following set of equations:

[0155] a+b=1

[0156]

[0157] In this embodiment, Gaussian distribution, as a continuous probability distribution, can better fit this randomness. Therefore, Gaussian distribution is used to represent the probability density distribution function of the operating state parameters of the nuclear reactor, which can be applied to the nuclear power field to obtain accurate target operation decisions to ensure the safe and stable operation of the nuclear reactor, improve the safety and safety management level of the nuclear reactor, and significantly reduce labor costs.

[0158] In one embodiment, the weight of the utility variable is positively correlated with the degree of influence of the operating state parameter corresponding to the utility variable on the fault.

[0159] It is understandable that, according to the importance of each operating state parameter in the nuclear reactor operation decision, a corresponding weight is assigned to the utility variable corresponding to the operating state parameter. The importance can be determined according to the actual situation, and the weight can be set according to the actual situation, and is not particularly limited here.

[0160] Weights are used to measure the importance of utility variables in alternative operating decisions, and the choice of weights will change the resulting target operating decision.

[0161] In this embodiment, the weight of the utility variable is positively correlated with the degree of influence of the operating status parameter corresponding to the utility variable on the fault, avoiding decision-making deviation caused by subjective bias or improper weight distribution, thereby ensuring the safe and stable operation of the nuclear reactor and improving the safety and safety management level of the nuclear reactor.

[0162] In one embodiment, step S103 includes: for each candidate operation decision, performing weighted summation on the actual utility values ​​of the utility variables, calculating the composite utility of the candidate operation decisions based on the weighted result, and determining the candidate operation decision with the largest composite utility as the target operation decision.

[0163] When evaluating multiple alternative operation decisions, first calculate the actual utility value of each utility variable in each alternative operation decision, then multiply the actual utility value by the corresponding weight, and obtain the composite utility of the alternative operation decisions through weighted summation to determine the target operation decision. Among them, the alternative operation decision with the largest composite utility is the target operation decision.

[0164] See also Figure 8 , the calculation formula of compound utility is as follows:

[0165]

[0166] Among them, U represents the composite utility, N represents the total number of utility variables, ω i represents the weight corresponding to each utility variable, x i represents the i-th utility variable, u(x i ) represents the value of the dependent variable of the utility function, that is, the actual utility value of the utility variable.

[0167] In this embodiment, for each alternative operating decision, the actual utility values ​​of the utility variables are weighted and summed, and the composite utility of the alternative operating decisions is calculated based on the weighted results. A unique target operating decision is determined based on the composite utility, thereby improving the safety and safety management level of the nuclear reactor and significantly reducing labor costs.

[0168] In one embodiment, step S103 includes: for each candidate operation decision, performing weighted summation on the actual utility values ​​of the utility variables, calculating the composite utility of the candidate operation decisions based on the weighted result, and determining the candidate operation decision with the largest composite utility as the target operation decision.

[0169] In this embodiment, for each alternative operating decision, the actual utility values ​​of the utility variables are weighted and summed, the composite utility of the alternative operating decisions is calculated based on the weighted results, and the target operating decision is determined according to the composite utility, thereby improving the safety and safety management level of the nuclear reactor and significantly reducing labor costs.

[0170] In one embodiment, the concept of "risk guidance" is introduced to couple the probabilistic decision-making process with the deterministic decision-making process for analysis, and each alternative operation decision is a sequence of decision trees. The step of calculating the composite utility of the alternative operation decision based on the weighted result includes: determining the success probability of each alternative operation decision according to the decision tree. Multiplying the success probability with the weighted result to obtain the composite utility of the alternative operation decision.

[0171] The goal of the deterministic decision analysis module is to integrate the current operating state parameters and the expected operating state parameters of the nuclear reactor and incorporate them into the decision-making process.

[0172] Probabilistic safety assessment (PSA) technology can realize probabilistic decision-making process. Compared with the deterministic decision-making process realized only by deterministic methods, probabilistic safety analysis technology can reveal the inherent risks of nuclear reactor operation in a wider range and provide more comprehensive and in-depth risk information. By applying probabilistic safety analysis technology, it is possible not only to quantify uncertainty and make comprehensive decisions guided by risks, but also to further improve the safety and safety management level of nuclear power plants.

[0173] However, the probabilistic decision-making process simply selects the alternative operation decision with the highest probability of success without considering other factors. For example, among the alternative operation decisions to avoid triggering a nuclear reactor shutdown, the one with the highest probability of success is manually shutting down the reactor, but if the changes in the operating state parameters of the nuclear reactor (such as temperature, etc.) are considered, this alternative operation decision may be determined as the most unfavorable alternative operation decision. In addition, the probabilistic decision-making process cannot explain the impact of the control behavior in the alternative operation decision on the operating state parameters, for example, executing the control behavior will have the effect of lowering the temperature and reducing the heat.

[0174] Because although the probabilistic decision analysis module automatically generates a corresponding set of control behaviors for each sequence, it does not specify the performance parameters of the components associated with the control behaviors. In addition, some instructions generated by the probabilistic decision analysis module only contain abstract action concepts without specific instructions. For example, an instruction may be to reduce power, but it does not specify how much power needs to be reduced.

[0175] For details, see Fig. 9 , through the probabilistic decision analysis module, the probabilistic decision analysis process is realized to obtain multiple alternative operation decisions, each of which is a sequence of the decision tree. Each alternative operation decision includes a different number of control behaviors. Based on the component status and availability corresponding to the control behavior, the success probability of each sequence of the decision tree is determined, that is, the success probability of each alternative operation decision is determined, and then the alternative operation decisions are screened based on the success probability. For example, if the component has not failed or the performance has not degraded, the success probability is high.

[0176] Fig. 9 The decision node in is the starting decision process, alternatives 1-N are N alternative operation decisions, actions 1-i, 1-j, 1-k and 1-l represent multiple control behaviors in each alternative operation decision, and the success probability p1-p Nis the success probability of each alternative operation decision, and the probabilistic decision analysis module can output the success probability corresponding to each alternative operation decision.

[0177] After the alternative operation scheme and the corresponding success probability are obtained in the probabilistic decision-making process, the alternative operation scheme and the success probability are transmitted to the deterministic decision analysis module. The deterministic decision process is realized by the deterministic decision analysis module. Specifically, the operation state parameters are selected, and the corresponding relationship between the operation state parameters and the utility variables in the multiple alternative operation decisions of the nuclear reactor is established, that is, the operation state parameters and the utility variables are linked to obtain the utility variables. The weight corresponding to the utility variable is determined, and the utility function is constructed. Finally, the actual utility value is obtained according to the utility function, and the composite utility is calculated according to the actual utility value and weight of the utility variable. Based on the composite utility, the target operation decision of the nuclear reactor is determined from the multiple alternative operation schemes obtained from the probabilistic decision process.

[0178] Taking into account the success probability of each control behavior of the alternative operation scheme, another calculation formula for the composite utility is obtained:

[0179]

[0180] Among them, U i represents the compound utility, p i represents the success probability of the ith alternative operation decision, that is, the success probability of the ith sequence, N represents the total number of utility variables, ω j represents the weight corresponding to each utility variable, x j represents the jth utility variable, u j (x j ) represents the value of the dependent variable of the utility function after the utility variable is substituted into the utility function, that is, the actual utility value of the utility variable.

[0181] In this embodiment, probabilistic methods and deterministic methods are combined for comprehensive decision-making to implement probabilistic decision-making processes and deterministic decision-making processes. Not only defense in depth and safety margin are taken into consideration, but also the success probability of alternative operating decisions is taken into consideration. Risk guidance is made, and risk awareness is integrated into the safety management model of nuclear reactors, thereby improving the safety and safety management level of nuclear reactors, significantly reducing labor costs, and alleviating unnecessary burdens.

[0182] In one embodiment, step S101 includes: screening out alternative operation decisions according to the success probability of each sequence of the decision tree; each sequence represents an operation decision; and the success probability is related to the expected loss.

[0183] According to the real-time status information of nuclear power plant equipment, the success probability of each sequence is determined and the sequences are sorted according to the component status and availability of the components associated with the control behavior. The probabilistic decision analysis module automatically selects alternative operation decisions as the best alternative operation decisions for executing the corresponding corrective measures. The selected decision alternatives comprehensively consider the uncertainty of component status and availability under the nuclear reactor failure state. Any option should be selected to prevent or minimize the possibility of starting the reactor protection system, so that the operating status parameters of the nuclear reactor are maintained within an acceptable range, so that the nuclear reactor can continue to operate normally.

[0184] In one embodiment, step S101 further includes: updating the success probability of the alternative operation decision in the decision tree according to the system equipment status of the nuclear reactor monitored in real time.

[0185] In one embodiment, step S101 further includes: arranging the sequences in reverse order according to the success probability, and determining the operation decisions corresponding to the sequences with success probabilities greater than a probability threshold as the alternative operation decisions; or, determining the operation decisions corresponding to a preset number of sequences ranked at the top as the alternative operation decisions.

[0186] The probability threshold can be set according to actual conditions and is not particularly limited here.

[0187] The use of risk-guided decision analysis methods, that is, combining probabilistic methods with deterministic methods for comprehensive decision-making, has the following advantages: (1) comprehensive consideration of events that pose potential challenges to nuclear power plants; (2) prioritization of these events that pose potential challenges to nuclear power plants; (3) more extensive consideration and integration of various resources to prevent or mitigate potential risks and challenges; (4) clear identification and quantification of sources of uncertainty; and (5) providing methods for sensitivity testing of the results of key assumptions, thereby making better decisions.

[0188] In this embodiment, the alternative operation decisions are arranged in reverse order according to the success probability, and the alternative operation decisions with a success probability greater than a probability threshold are screened, which greatly reduces the manpower cost and alleviates unnecessary burden.

[0189] In one embodiment, after the step of determining the success probability of each alternative operation decision according to the decision tree, the nuclear reactor risk-guided multi-attribute autonomous operation decision method also includes: arranging the alternative operation decisions in reverse order according to the success probability, and only retaining the alternative operation decision with the highest success probability.

[0190] According to the real-time status information of nuclear power plant equipment, the success probability is determined according to the component status and availability of the components corresponding to the alternative operation decisions, and multiple alternative operation decisions are ranked. The probabilistic decision analysis module automatically selects the best candidate control option, that is, only the alternative operation decision with the highest probability of success is retained.

[0191] In this embodiment, the success probability of alternative operating decisions is taken into consideration, risk guidance is made, and only the alternative operating plans with the highest success probability are retained. Risk awareness is integrated into the safety management model of the nuclear reactor, the safety and safety management level of the nuclear reactor is improved, the labor cost is greatly reduced, the possibility of starting the nuclear reactor protection system is minimized, and the various operating status parameters of the nuclear reactor are maintained within a safe range, so that the nuclear reactor can continue to operate normally.

[0192] The failure scenario of a steam turbine control valve is selected below to verify the effectiveness of the risk-guided multi-attribute autonomous operation decision-making method for nuclear reactors.

[0193] The turbine control valve is an important control element in a nuclear reactor. The opening of the turbine control valve directly affects the steam flow of the turbine. When the opening of the turbine control valve decreases and moves toward the closing direction, the steam flow of the turbine will decrease, which will in turn affect the cooling flow of the steam generator. The opening of the feedwater flow control valve increases, which increases the cooling flow of the steam generator, resulting in overcooling of the main circuit system; conversely, the opening of the feedwater flow control valve decreases, which reduces the cooling flow of the steam generator, resulting in undercooling of the main circuit system. If the steam flow cannot be increased or the feedwater flow cannot be reduced, it will cause heat imbalance in the secondary cooling system and shutdown of the nuclear reactor.

[0194] After determining the fault scenario, a nuclear reactor risk-guided multi-attribute autonomous operation decision method is used to determine the target operation decision.

[0195] Alternative operating plans include, but are not limited to: moving the turbine control valve in the closing direction, resetting the turbine control valve, keeping the steam generator bypass valve in the open state, adjusting the turbine bypass valve or feed water flow, reducing the feed water flow to the steam generator, opening the turbine bypass valve, reducing the power of the nuclear reactor, closing the turbine bypass valve and manually shutting down the nuclear reactor, etc. The specific options can be selected according to actual conditions and are not particularly limited here.

[0196] After screening the alternative operation plans, the following five alternative operation plans are obtained:

[0197] 1) The nuclear reactor is shut down when the water level in the steam generator is low, i.e. no measures are taken;

[0198] 2) Reset the turbine control valve to restore it to its normal working position;

[0199] 3) Open the turbine bypass valve for short-term compensation; it is also recommended to reduce the power of the nuclear reactor or repair the fault of the turbine control valve.

[0200] 4) If the power of a nuclear reactor does not reach 100%, open the steam generator bypass valve of the nuclear reactor; at the same time, it is recommended to manually shut down another nuclear reactor or repair the fault of the steam turbine control valve;

[0201] 5) Reduce the feed water flow to the steam generator; it is also recommended to reduce the power of the nuclear reactor or repair the failure of the turbine control valve.

[0202] The above five alternative operation decisions are the five sequences of the decision tree. The success probability of each alternative operation decision is determined according to the decision tree, and the alternative operation decisions are arranged according to the success probability. The sorted order is shown in Table 4.

[0203] Table 4 Control options sorted by likelihood of success

[0204]

[0205] Among them, the possibility of success is the probability of success, the control option is the alternative operation decision, and the reactor module 1 is a nuclear reactor. The first row of Table 4 corresponds to the first of the five alternative operation decisions, that is, no measures are taken. The second row of Table 4 corresponds to the fourth of the five alternative operation decisions, that is, manually shutting down another nuclear reactor or repairing the fault of the turbine control valve. The third row of Table 4 corresponds to the fifth of the five alternative operation decisions, that is, reducing the feedwater flow to the steam generator; at the same time, it is recommended to reduce the power of the nuclear reactor or repair the fault of the turbine control valve. The fourth row of Table 4 corresponds to the third of the five alternative operation decisions, that is, opening the turbine bypass valve for short-term compensation. The fifth row of Table 4 corresponds to the second of the five alternative operation decisions, that is, resetting the turbine control valve to restore the turbine control valve to its normal working position.

[0206] It can be seen that although the success probability of the first alternative operation decision is very high, the result is an emergency shutdown, which means that the nuclear reactor will stop operating immediately. Although this can avoid potential safety risks, it will also bring significant economic losses and production interruptions. Therefore, although the first alternative operation decision has a high probability of success, it is usually not the first choice in actual applications. The second alternative operation decision is to manually shut down the nuclear reactor. Although it helps to stabilize the system, it will cause the nuclear reactor power to drop to 50%, which means that the power generation capacity of the nuclear reactor will drop significantly. In addition, shutting down a nuclear reactor may also affect the load balancing and stability of the entire nuclear power plant. Although the fifth alternative operation decision has a low probability of success, once it succeeds, the nuclear reactor will be able to resume 100% power operation. However, due to the low probability of success, it is risky in actual applications.

[0207] The third and fourth alternative operation decisions both have a high probability of success, and the results are both power reduction. Although the specific degree of power reduction is not given, the third and fourth alternative operation decisions attempt to balance heat and flow by adjusting the operating parameters to maintain the stability of the nuclear reactor. Compared with emergency shutdown or significant power reduction, the third and fourth alternative operation decisions are more moderate and avoid direct major economic losses. Since the results of the third and fourth alternative operation decisions are relatively abstract, the results are not specifically stated. Therefore, in order to more accurately evaluate the pros and cons of the third and fourth alternative operation decisions, the third and fourth alternative operation decisions are selected and transmitted to the deterministic decision analysis module for analysis to obtain accurate target operation decisions.

[0208] Through the input of the variable sensor, the change of the operating state parameters after the execution of the two alternative operating decisions can be obtained. In this embodiment, the operating state parameters include but are not limited to the following four: core outlet temperature, main circuit cold pool temperature, steam header pressure and steam drum liquid level. The operating state parameters can be selected according to the actual situation and are not particularly limited here. The operating state parameters can usually be used to characterize the physical behavior of the nuclear reactor, or to predict whether the nuclear reactor will fail in the future.

[0209] Taking the third alternative operation decision as an example, after executing the third alternative operation decision, the change of the core outlet temperature over time is as follows: Fig.10 As shown in . Among them, reactor module 1 and reactor module 2 are two nuclear reactors, the horizontal axis represents time, and the vertical axis represents the core outlet temperature. After executing the third alternative operation decision, the utility value corresponding to the core outlet temperature changes as shown in Fig.11As shown. Among them, the red circle marks the boundary point of the safety range. The boundary point can be used to determine whether the nuclear reactor is shut down. The specific setting can be based on the actual situation and is not particularly limited here. The actual utility value of the core outlet temperature of reactor module 2 is positive, while the actual utility value of the core outlet temperature of reactor module 1 is negative. The closer to the boundary value, the closer the actual utility value of the utility variable is to 0, and the magnitude of the decrease in the actual utility value will increase rapidly. For the nuclear reactor control system, the actual utility value is negative, indicating that the alternative operation decision cannot be the target operation decision.

[0210] Table 5 lists the parameter values ​​of the operating state parameters of the nuclear reactors for the third alternative operating decision and the fourth alternative operating decision, wherein RX1 represents the first nuclear reactor, that is, reactor module 1; RX2 represents the second nuclear reactor, that is, reactor module 2; control option 3 represents the third alternative operating decision; and control option 4 represents the fourth alternative operating decision.

[0211] Table 5 Attribute values ​​of key process variables for reactor modules with control option 3 and control option 4

[0212]

[0213] The corresponding relationship between the operating state parameters of the nuclear reactor and the utility variables is established. Specifically, the operating state parameters are converted into utility variables by using the standard 0-1 transformation, and the obtained utility variables are shown in Table 6.

[0214] Table 6 Utility variable matrix

[0215]

[0216] For each alternative operation decision, the weight corresponding to the utility variable is obtained, and the utility function is constructed for the operation state parameter. Specifically, the utility function corresponding to each utility variable is obtained through affine transformation, and the transformation coefficients a and b of the affine transformation are shown in Table 7.

[0217] Table 7 The corresponding transformation coefficients a and b in the utility function of each utility variable

[0218]

[0219] The utility functions of various utility variables are as follows: Fig.12 The utility function represents the corresponding relationship between the parameter value of the operating state parameter and the actual utility value of the utility variable. The actual utility value of the utility variable corresponding to the operating state parameter whose parameter value does not exceed the safety range is greater than the actual utility value of the utility variable corresponding to the operating state parameter whose parameter value exceeds the safety range.

[0220] The actual utility values ​​of the utility variables corresponding to the multiple alternative operation decisions are calculated.

[0221] Table 8 lists the actual utility values ​​of utility variables, reactor modules and total utility values ​​of the power plant, and calculates the composite utility based on the success probability and the actual utility value. Among them, the success probability is the success probability.

[0222] Table 8 Utility values ​​of control options 3 and 4

[0223]

[0224] The actual utility values ​​of the utility variables are weighted and summed, and the composite utility of the alternative operation decision is calculated based on the weighted result. The success probability is multiplied by the weighted result to obtain the composite utility of the alternative operation decision.

[0225] In this embodiment, the nuclear reactor risk-guided multi-attribute autonomous operation decision-making method is effectively applied. Although the success probability of the third alternative operation decision is slightly higher than that of the fourth alternative operation decision, the composite utility of the fourth alternative operation decision is significantly higher than that of the third alternative operation decision. The fourth alternative operation decision is a more ideal decision-making scheme, which achieves the goals of minimizing risk and optimizing performance. The nuclear reactor risk-guided multi-attribute autonomous operation decision method is of great significance for guiding the operation and maintenance of the actual nuclear power field, and also provides a useful reference for similar decision-making problems in the future.

[0226] This research result is practical in the field of autonomous operation decision-making of nuclear reactors, and provides new ideas and methods for improving the quality and efficiency of engineering decision-making.

[0227] This embodiment introduces utility theory into the engineering field. This attempt not only broadens the scope of application of utility theory, but also provides a new perspective for solving decision-making problems in the engineering field. Utility theory was originally mainly used in the field of economics to describe and predict the decision-making behavior of individuals when faced with multiple choices. By constructing a utility function, the theory can quantify the degree of satisfaction or utility brought to individuals by different choices, thereby helping individuals make optimal decisions. In economics, the application of utility theory has been quite mature, providing strong support for resource allocation, market analysis and policy making. However, in the engineering field, decision-making problems of large and complex control systems often involve multiple interrelated factors and complex constraints, which makes traditional optimization methods difficult to deal with. This embodiment introduces utility theory into the engineering field, aiming to construct a utility function suitable for engineering problems. The risk-guided multi-attribute autonomous operation decision-making method for nuclear reactors makes the decision results more in line with the actual needs and safety standards of nuclear power plants, not only overcomes the limitations of traditional methods, but also makes the solutions to engineering problems more in line with actual conditions and the needs of decision makers.

[0228] Specifically, firstly, it broadens the scope of application of utility theory, introduces it from the field of economics to the field of engineering, provides new theoretical support for the solution of engineering problems and expands the definition of utility function; secondly, by constructing utility functions and decision analysis methods suitable for engineering problems, it improves the accuracy and effectiveness of solutions to engineering problems; thirdly, it provides decision makers with more comprehensive and flexible decision-making tools, which helps to improve the scientificity and rationality of decisions.

[0229] In this embodiment, the probabilistic decision analysis module is combined with the deterministic decision analysis module to give full play to the advantages of both. By selecting the fault scenario of the turbine control valve opening decreasing and moving in the closing direction as a case analysis, the effectiveness of the proposed method in actual engineering problems is verified.

[0230] In summary, the risk-guided multi-attribute autonomous operation decision-making method for nuclear reactors not only enriches the application of multi-attribute utility theory in the field of engineering decision-making, but also provides new ideas and methods for risk-guided comprehensive decision-making analysis methods. It is innovative and practical in the engineering field, and is of great significance for improving the quality and efficiency of engineering decision-making.

[0231] Example 2

[0232] Corresponding to the aforementioned nuclear reactor risk-guided multi-attribute autonomous operation decision-making method embodiment, the present disclosure also provides an embodiment of a nuclear reactor risk-guided multi-attribute autonomous operation decision-making system.

[0233] Fig.13 A schematic diagram of a module of a nuclear reactor risk-guided multi-attribute autonomous operation decision system provided by an exemplary embodiment of the present disclosure, wherein the nuclear reactor operation decision analysis system based on multi-attribute utility theory includes:

[0234] A probabilistic decision analysis module 201 is used to establish a plurality of alternative operation decisions of the nuclear reactor, wherein the alternative operation decisions include an expected loss of the nuclear reactor after executing the control behavior included in the alternative operation decision, wherein the expected loss satisfies a preset loss condition;

[0235] The deterministic decision analysis module 202 is used to simulate the nuclear reactor based on each alternative operation decision to obtain the expected parameter value of the operation state parameter of the nuclear reactor and the expected utility value of the corresponding effect variable during the simulation process, and substitute the corresponding expected utility value into the utility function to calculate the actual utility value of each utility variable; wherein the utility function represents the corresponding relationship between the operation state parameter and the utility variable, and the actual utility value of the utility variable corresponding to the state parameter whose parameter value does not exceed the safety range is greater than the actual utility value of the utility variable corresponding to the state parameter whose parameter value exceeds the safety range;

[0236] The optimal decision generating module 203 is used to determine the target operation decision of the nuclear reactor from a plurality of alternative operation decisions based on the actual utility value.

[0237] Among them, the best decision is the target operation decision.

[0238] Optionally, the utility function is obtained by the following steps: for each operating state parameter, the utility function is obtained by performing an affine transformation on a probability density distribution function of the operating state parameter.

[0239] Optionally, the probability density distribution function is represented by Gaussian distribution.

[0240] Optionally, the probabilistic decision analysis module 201 includes:

[0241] A decision screening unit, used to screen out the alternative operation decisions according to the success probability of each sequence of the decision tree; each sequence represents an operation decision; the success probability is related to the expected loss;

[0242] and / or, a success probability updating unit, configured to update the success probability of the alternative operation decision in the decision tree according to the system equipment status of the nuclear reactor monitored in real time;

[0243] and / or, a sorting unit, used to sort the sequence in reverse order according to the success probability, and determine the operation decision corresponding to the sequence with a success probability greater than a probability threshold as an alternative operation decision;

[0244] Alternatively, the sorting unit is used to determine the operation decisions corresponding to a preset number of sequences that are ranked first as the candidate operation decisions.

[0245] Optionally, the optimal decision generating module 203 includes:

[0246] A weight determination unit, used to determine the weight corresponding to the utility variable;

[0247] A weighted summing unit, for performing weighted summing of actual utility values ​​of utility variables for each alternative operation decision, and calculating a composite utility of the alternative operation decision based on the weighted result;

[0248] The target operation decision determination unit is used to determine the alternative operation decision with the largest composite utility as the target operation decision.

[0249] Optionally, the weighted summation unit is specifically used to: determine the success probability of each alternative operation decision according to a decision tree; wherein each alternative operation decision is a sequence of the decision tree; and multiply the success probability by the weighted result to obtain the composite utility of the alternative operation decision.

[0250] In this embodiment, a multi-attribute utility theory algorithm is used as a mathematical analysis tool, which can comprehensively consider multiple operating status parameters, measure the value of multiple alternative operating decisions by establishing a unified scale, or measure the decision maker's satisfaction with multiple alternative operating decisions, realize cross-standard comparison and decision-making, calculate the actual utility value of the utility variable, quantify the impact of different alternative operating decisions on the nuclear reactor through the actual utility value, and obtain accurate target operating decisions. It is possible to make fast and accurate decisions in a complex and changing environment to ensure the safe and stable operation of the nuclear reactor, improve the safety and safety management level of the nuclear reactor, greatly reduce labor costs, improve the scientificity and accuracy of decision-making, reduce decision-making risks, and provide strong support for engineering practice.

[0251] As for the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The system embodiment described above is only illustrative, wherein the units described as separate components may or may not be physically separated, and the components as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the disclosed solution.

[0252] Example 3

[0253] Fig.14 This is a schematic diagram of the structure of an electronic device shown in an example embodiment of the present disclosure, the electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor, and when the processor executes the computer program, the nuclear reactor risk-guided multi-attribute autonomous operation decision method described in any of the above embodiments is implemented. Fig.14 The electronic device 30 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0254] like Fig.14 As shown, the electronic device 30 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including the memory 32 and the processor 31).

[0255] The bus 33 includes a data bus, an address bus, and a control bus.

[0256] The memory 32 may include a volatile memory, such as a random access memory (RAM) 321 and / or a cache memory 322 , and may further include a read-only memory (ROM) 323 .

[0257] The memory 32 may also include a program tool 325 (or utility) having a set (at least one) of program modules 324, such program modules 324 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0258] The processor 31 executes various functional applications and data processing by running the computer program stored in the memory 32, such as the nuclear reactor risk-guided multi-attribute autonomous operation decision method provided in any of the above embodiments.

[0259] The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 35. Furthermore, the electronic device 30 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 36. As shown, the network adapter 36 communicates with other modules of the electronic device 30 via a bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0260] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0261] Example 4

[0262] The embodiments of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the nuclear reactor risk-guided multi-attribute autonomous operation decision-making method provided in any of the above embodiments is implemented.

[0263] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0264] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, but these changes and modifications all fall within the protection scope of the present disclosure.

Claims

1. A nuclear reactor risk-guided multi-attribute autonomous operation decision-making method, characterized in that: The nuclear reactor risk-guided multi-attribute autonomous operation decision-making method comprises: Establishing a plurality of alternative operation decisions for a nuclear reactor, wherein the alternative operation decisions include expected losses of the nuclear reactor after executing control actions included in the alternative operation decisions, wherein the expected losses meet preset loss conditions; The nuclear reactor is simulated based on each alternative operation decision to obtain expected parameter values ​​of the operation state parameters of the nuclear reactor and expected utility values ​​of the corresponding effect variables during the simulation process, and the corresponding expected utility values ​​are substituted into the utility function to calculate the actual utility values ​​of the utility variables; wherein the utility function represents the corresponding relationship between the operation state parameters and the utility variables, and the actual utility value of the utility variable corresponding to the state parameter whose parameter value does not exceed the safety range is greater than the actual utility value of the utility variable corresponding to the state parameter whose parameter value exceeds the safety range; Based on the actual utility value, a target operating decision for the nuclear reactor is determined from a plurality of alternative operating decisions.

2. The nuclear reactor risk-guided multi-attribute autonomous operation decision-making method according to claim 1, characterized in that: The utility function is obtained by the following steps: For each of the operating status parameters, a utility function is obtained by performing an affine transformation on the probability density distribution function of the operating status parameter.

3. The nuclear reactor risk-guided multi-attribute autonomous operation decision-making method according to claim 2, characterized in that: The probability density distribution function is represented by Gaussian distribution.

4. The nuclear reactor risk-guided multi-attribute autonomous operation decision-making method according to any one of claims 1 to 3, characterized in that: The step of establishing a plurality of alternative operating decisions for a nuclear reactor comprises: According to the success probability of each sequence of the decision tree, the alternative operation decisions are screened out; each sequence represents an operation decision; the success probability is related to the expected loss; And / or, based on the system equipment status of the nuclear reactor monitored in real time, the success probability of the alternative operation decision in the decision tree is updated; and / or, the sequence is arranged in reverse order according to the success probability, and the operation decision corresponding to the sequence with a success probability greater than a probability threshold is determined as the alternative operation decision; or, the operation decision corresponding to a preset number of sequences that are ranked first is determined as the alternative operation decision.

5. The nuclear reactor risk-guided multi-attribute autonomous operation decision-making method according to claim 4, characterized in that: The step of determining the target operation decision of the nuclear reactor from a plurality of alternative operation decisions based on the actual utility value comprises: determining a weight corresponding to the utility variable; For each of the alternative operation decisions, weighted summing the actual utility values ​​of the utility variables according to the weights, and calculating the composite utility of the alternative operation decisions based on the weighted results; The alternative operating decision with the maximum composite utility is determined as the target operating decision.

6. The nuclear reactor risk-guided multi-attribute autonomous operation decision-making method according to claim 5, characterized in that: The step of calculating the composite utility of the alternative operation decision based on the weighted results includes: The success probability is multiplied by the weighted result to obtain the composite utility of the alternative operation decision.

7. The nuclear reactor risk-guided multi-attribute autonomous operation decision-making method according to any one of claims 1 to 6, characterized in that: The effect variable is obtained by the following formula: Among them, x i is the i-th utility variable, p i is the parameter value of the i-th operating state parameter; (p i ) max is the maximum value of the i-th operating state parameter, (p i ) min is the minimum value of the i-th operating state parameter; And / or, the weights are as follows: Where N represents the total number of utility variables, ω j represents the weight corresponding to the j-th utility variable; And / or, the probability density distribution function is expressed as follows: Wherein, p(x|μ,σ) represents the probability density distribution function, μ represents the mean value of the probability density distribution, σ represents the standard deviation of the probability density distribution, and e represents the base of the natural logarithm; And / or, the utility function is expressed as follows: Wherein, u(x|μ,σ) represents the utility function, and a and b are transformation coefficients of the affine transformation; the transformation coefficients a and b of the affine transformation are calculated by the following formula: a+b=1 And / or, the compound utility is obtained by the following formula: Wherein, U represents the composite utility, and u(xi) represents the actual utility value of the utility variable.

8. A nuclear reactor risk-guided multi-attribute autonomous operation decision system, characterized in that: The nuclear reactor operation decision analysis system based on the multi-attribute utility theory includes: A probabilistic decision analysis module, used to establish a plurality of alternative operation decisions for a nuclear reactor, wherein the alternative operation decisions include an expected loss of the nuclear reactor after executing the control behavior included in the alternative operation decision, and the expected loss satisfies a preset loss condition; A deterministic decision analysis module is used to simulate the nuclear reactor based on each alternative operation decision to obtain the expected parameter value of the operation state parameter of the nuclear reactor and the expected utility value of the corresponding effect variable during the simulation process, and substitute the corresponding expected utility value into the utility function to calculate the actual utility value of each utility variable; wherein the utility function represents the corresponding relationship between the operation state parameter and the utility variable, and the actual utility value of the utility variable corresponding to the state parameter whose parameter value does not exceed the safety range is greater than the actual utility value of the utility variable corresponding to the state parameter whose parameter value exceeds the safety range; The optimal decision generating module is used to determine the target operation decision of the nuclear reactor from a plurality of alternative operation decisions based on the actual utility value.

9. The nuclear reactor risk-guided multi-attribute autonomous operation decision system according to claim 8, characterized in that: The effect variable is obtained by the following formula: Wherein, xi is the i-th utility variable, pi is the parameter value of the i-th operating state parameter; (pi)max is the maximum value of the i-th operating state parameter, and (pi)min is the minimum value of the i-th operating state parameter; And / or, the weights are as follows: Wherein, N represents the total number of utility variables, ωj represents the weight corresponding to the jth utility variable; And / or, the probability density distribution function is expressed as follows: Wherein, p(x|μ,σ) represents the probability density distribution function, μ represents the mean value of the probability density distribution, σ represents the standard deviation of the probability density distribution, and e represents the base of the natural logarithm; And / or, the utility function is expressed as follows: Wherein, u(x|μ,σ) represents the utility function, and a and b are transformation coefficients of the affine transformation; the transformation coefficients a and b of the affine transformation are calculated by the following formula: a+b=1 And / or, the compound utility is obtained by the following formula: Wherein, U represents the composite utility, and u(xi) represents the actual utility value of the utility variable.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the nuclear reactor risk-guided multi-attribute autonomous operation decision method according to any one of claims 1 to 6 is implemented.

Images