Intelligent assessment method and system for generator set supervision personnel

Abstract

The invention relates to the technical field of assessment management, in particular to an intelligent assessment method and system for generator set supervision personnel. The system comprises an examinee face feature image data acquisition module, a cheating behavior judgment module, a generator set equipment operation data acquisition module, an error interval setting module, an examinee reply result data acquisition module and an examinee reply result scoring module. Generator set equipment is randomly selected as examination question equipment for examinees to reply, examination question equipment operation parameter data replied by the examinees are collected, and examination question reply results are scored in combination with an error interval matrix, so that errors caused by manual auditing are avoided, the consumption of human resources is reduced, and meanwhile, the accuracy of the scoring results is ensured; through a beetle beetle algorithm, image matching of examinee face feature image data and standard face feature image data is realized, the accuracy of image matching is ensured, the possibility of cheating behaviors is effectively avoided, and intelligent supervision in an examination process is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of assessment management, and in particular to an intelligent assessment method and system for generator set monitoring personnel. Background Art

[0002] A generator set is composed of multiple devices with different functions, and its specific function is to convert other forms of energy into electrical energy.

[0003] During the operation of the generator set, the operating data of the unit needs to be monitored in real time. This process is often achieved by monitoring personnel. If the abnormalities that occur during the operation of the generator set are not discovered and corrected by the monitoring personnel in time, huge harm and losses are likely to occur.

[0004] The Chinese patent invention with publication number CN113283820B introduces an assessment and evaluation system for station attendants, including: a central server, used to generate and send assessment questions, the assessment questions include multiple assessment contents; a forwarding server, used to receive the assessment questions sent by the central server, and correspondingly forward the multiple assessment contents included in the assessment questions; multiple assessment terminals, used to receive the assessment contents corresponding to themselves sent by the forwarding server, and assess the assessment objects according to the assessment contents, so as to improve the consistency of the assessment process with the actual business processing process, but the assessment standards are not set, and the accuracy of the scoring results cannot be guaranteed. At the same time, it is difficult to prevent cheating that may occur during the assessment process. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In order to solve the deficiencies in the background technology, the present invention proposes an intelligent assessment method and system for generator set monitoring personnel, which realizes the intelligence of monitoring spot checks, effectively eliminates the possibility of cheating, and realizes intelligent supervision during the examination process.

[0007] (II) Technical solution

[0008] An intelligent assessment method for generator panel monitors, which tracks whether candidates have cheated during the assessment process and scores the candidates' responses; the method includes the following steps:

[0009] Collect image data during the examinee's answering process and perform preprocessing to obtain the examinee's facial feature image dataset;

[0010] Establish a standardized facial feature image data set, match each candidate's facial feature image data in the candidate's facial feature image data set with each standardized facial feature image data in the standardized facial feature image data set in turn, and determine whether the candidate has cheated based on the matching results;

[0011] Setting a generator set equipment set and an equipment operation parameter category set, collecting operation data of the generator set equipment during operation according to the generator set equipment set and the equipment operation parameter category set, and obtaining a generator set equipment operation data matrix;

[0012] The error interval of each generator set equipment operation data in the generator set equipment operation data matrix is ​​set by an intelligent optimization algorithm to obtain an error interval matrix;

[0013] Randomly select the generator set equipment as the test equipment to obtain the test equipment set, collect the test taker's answer result data, and obtain the test taker's answer result data matrix;

[0014] The candidate response results are scored using the error interval matrix and the candidate response result data matrix.

[0015] The present invention obtains facial feature image data of examinees by preprocessing image data collected by an image collection device at irregular intervals, thereby ensuring that the collected image data are random and universal; sequentially matching facial feature image data of each examinee with all standard facial feature image data in a standard facial feature image data set, and judging whether the examinee has cheated according to the matching result, thereby ensuring the reliability and accuracy of the judgment result; setting a corresponding error interval matrix for a generator set equipment operation data matrix by a firefly algorithm, thereby reducing the error of the setting process; and constructing a mathematical model by simulating the behavior that fireflies are attracted by fireflies with high brightness, thereby improving the speed and efficiency of the optimization process, and effectively preventing the optimization result from falling into a local optimum; and randomly selecting a generator set equipment as a test question equipment, allowing the examinee to answer according to the test question equipment, collecting the examinee's answer result data, and scoring the examinee's answer result in combination with the error interval matrix, thereby avoiding the error caused by manual review, reducing the consumption of human resources, and ensuring the accuracy of the scoring result.

[0016] Preferably, the specific steps of collecting image data during the examinee's answering process and preprocessing to obtain the examinee's facial feature image data set are as follows:

[0017] The image data of the examinee's answering process is collected from time to time by the image acquisition device installed on the display screen used by the examinee to answer, and the examinee's face image set E = {e 1 ,e 2 ,…,e i ,…,e p}, where e i represents the i-th examinee face image collected by the image collection device, and p represents the total number of examinee face images collected;

[0018] The sliding average filter is used to perform low-pass filtering on each candidate's face image in the candidate's face image set to obtain the candidate's face feature image data set E' = {e' 1 ,e' 2 ,…,e' i ,…,e' p}, where e' i It represents the facial feature image data of the examinee obtained after the i-th examinee's facial image is processed by low-pass filtering.

[0019] By preprocessing the image data collected by the image acquisition device at irregular intervals, the facial feature image data of the examinee is obtained, ensuring the randomness and universality of the collected image data.

[0020] Preferably, a standardized facial feature image data set is established, and each candidate's facial feature image data in the candidate's facial feature image data set is matched with each standardized facial feature image data in the standardized facial feature image data set in turn. The specific steps of determining whether the candidate has cheated according to the matching results are as follows:

[0021] Establish a standard facial feature image dataset G = {g 1 ,g 2 ,…,g i ,…,g u}, where g i represents the i-th standard facial feature image data, and u represents the total number of standard facial feature image data;

[0022] Perform image matching on each candidate's facial feature image data in the candidate's facial feature image data set with all the standard facial feature image data in the standard facial feature image data set in sequence;

[0023] The image matching comprises the following steps:

[0024] Construct a population of longhorn beetles, set the population size of longhorn beetles to N', the second current number of iterations to T, and the second maximum number of iterations to T max And the dimension of the search space of normalized facial feature image data is M;

[0025] Each individual longhorn beetle in the longhorn beetle population corresponds to a standard facial feature image data in the standard facial feature image data set, and the initial position set Y of the longhorn beetle population is randomly generated in the standard facial feature image data search space. 1 ,Y 2 ,…,Y i ,…,Y p}, where Y i represents the initial position of the i-th individual in the beetle population;

[0026] Calculate the fitness function value of each individual beetle in the beetle population, and take the individual beetle with the highest fitness function value as the current optimal individual; the fitness function formula is as follows:

[0027]

[0028] Among them, F(y i ) represents the fitness function value of the i-th individual in the beetle population, y i represents the degree of deviation between the standard facial feature image data corresponding to the i-th longicorn individual and the facial feature image data of the candidate being matched, and φ represents the second correction value;

[0029] The iteration operation begins. During each iteration, each individual beetle in the beetle population uses the two antennae on the left and right sides of its head to detect the concentration of food odor in the standard facial feature image data search space, selects a direction to update its position according to the concentration of the food odor, recalculates the fitness function value of each individual beetle after the position update, and selects the individual beetle with the highest fitness function value as the current optimal individual.

[0030] Determine whether the second current iteration number T is less than the second maximum iteration number T max , if the second current iteration number T is less than the second maximum iteration number T max , then the second current iteration number is T plus 1, and the iteration operation is performed again; otherwise, the current optimal individual is taken as the global optimal solution, and the fitness function threshold is set. When the fitness function value of the global optimal solution is greater than the fitness function threshold, it is determined that the standard facial feature image data corresponding to the global optimal solution matches the facial feature image data of the candidate to be matched successfully; otherwise, it is determined that the facial feature image data of the candidate to be matched and all the standard facial feature image data in the standard facial feature image data set are not matched successfully;

[0031] Count the total number of candidate facial feature image data that successfully matches the standard facial feature image data in the candidate facial feature image dataset σ, and calculate the standard value of the candidate facial image

[0032] Determine whether the examinee has cheated; including the following steps:

[0033] Set the standard threshold of the examinee's face image to υ;

[0034] If ω≥υ, it is determined that the examinee has not cheated, and the examinee's facial feature image data is saved in the examinee data management module;

[0035] If ω<υ, it is determined that the examinee may have cheated, and the examinee's facial feature image data is fed back to the main control unit for secondary determination. After the secondary determination is completed, the examinee's facial feature image data is saved in the examinee data management module.

[0036] By matching each candidate's facial feature image data with all the standard facial feature image data in the standard facial feature image data set in turn, whether the candidate has cheated is determined based on the matching results, thereby ensuring the reliability and accuracy of the determination results; by using the beetle whisker algorithm to achieve image matching between the candidate's facial feature image data and the standard facial feature image data, accurate matching of facial feature images is completed, effectively eliminating the possibility of cheating and realizing intelligent supervision during the examination.

[0037] Preferably, the specific steps of setting a generator set equipment set and an equipment operation parameter category set, collecting the operation data of the generator set equipment during operation according to the generator set equipment set and the equipment operation parameter category set, and obtaining the generator set equipment operation data matrix are as follows:

[0038] Set the generator set A = {a 1 ,a 2 ,…,a i ,…,a k}, where a i represents the i-th generator set, k represents the total number of generator sets;

[0039] Set the equipment operation parameter category set B = {b 1 ,b 2 ,…,b i ,…,b l}, where b i represents the i-th type of operating parameter in the working process of the generator set equipment, and l represents the total number of operating parameter categories in the working process of the generator set equipment;

[0040] According to the generator set A={a 1 ,a 2 ,…,a i ,…,a k} and the equipment operation parameter category set B = {b 1 ,b 2 ,…,b i ,…,b l The operating data of the generator set equipment during operation is collected to obtain the generator set equipment operating data matrix C as follows:

[0041]

[0042] Among them, c ijRepresents the data of the jth type of operating parameters during the operation of the i-th generator set.

[0043] Generator set equipment is randomly selected as the test equipment, and candidates are asked to answer according to the test equipment. The data of candidates' answers are collected to ensure the reliability and comprehensiveness of the answer results.

[0044] Preferably, the error interval of each generator set equipment operation data in the generator set equipment operation data matrix is ​​set by an intelligent optimization algorithm, and the specific steps of obtaining the error interval matrix are as follows:

[0045] Construct a firefly population, set the firefly population size to N, the light intensity absorption coefficient to γ, the step size perturbation factor to α, the first current iteration number to t, and the first maximum iteration number to t max and the dimension of the running parameter search space is P';

[0046] Based on any one of the generator set operation data in the generator set operation data matrix C, N operation parameter intervals are randomly generated. Each firefly individual in the firefly population represents an operation parameter interval. The initial position set of the firefly population is set to X = {X 1 ,X 2 ,…,X i ,…,X N}, where X i represents the initial position of the i-th firefly individual in the firefly population;

[0047] Calculate the fitness function value of each firefly individual in the firefly population, and take the firefly individual with the highest fitness function value as the current optimal individual; the fitness function value formula is as follows:

[0048]

[0049] Among them, Fit(x i ) represents the fitness function value of the i-th firefly individual in the firefly population, x i represents the effect of the i-th firefly individual corresponding to the operating parameter interval on maintaining the normal operation of the generator set equipment, and ε represents the first correction value;

[0050] Start iterating and calculate the spatial distance between individual fireflies in the firefly population; the spatial distance calculation formula is as follows:

[0051]

[0052] Among them, D i,j represents the spatial distance between the i-th firefly individual and the j-th firefly individual, X i,z and X j,zThey represent the positions of the i-th firefly individual and the j-th firefly individual in the z-th dimension operating parameter search space respectively;

[0053] In the operating parameter search space, firefly individuals with low fitness function values ​​are attracted to firefly individuals with high fitness function values; the attraction function formula is as follows:

[0054]

[0055] Among them, β(D i,j ) represents the attraction between the i-th firefly individual and the j-th firefly individual, β 0 It represents the attraction between two fireflies when the distance between them is 0;

[0056] Firefly individuals will be attracted by fireflies with high fitness function values ​​to update their positions in the operating parameter search space; the position update formula is as follows:

[0057]

[0058] in, and They represent the position of the i-th firefly individual in the firefly population during the t+1th and tth iterations, respectively. represents the position of the j-th firefly individual in the firefly population during the t-th iteration, and rand represents a random number between (0, 1);

[0059] Since the current best individual is the firefly individual with the highest fitness function value, the current best individual is set to randomly select a direction to update its position in the running parameter search space;

[0060] Calculate the fitness function value of each firefly individual in the firefly population after the position is updated, and take the firefly individual with the highest fitness function value as the current optimal individual;

[0061] Determine whether the first current iteration number t is less than the first maximum iteration number t max , if the first current iteration number t is less than the first maximum iteration number t max , then the first current iteration number t is increased by 1, and the iteration operation is performed again; otherwise, the current optimal individual is taken as the global optimal solution, and the operating parameter interval corresponding to the global optimal solution is output as the error interval of the operating data of the generator set equipment;

[0062] Repeat the above operation until the error intervals of all the generator set equipment operation data in the generator set equipment operation data matrix are obtained, and the error interval matrix Q is generated as follows:

[0063]

[0064] in, It represents the error interval of the jth type of operating parameters during the operation of the i-th generator set.

[0065] Compared with the method of calling historical data to set the error interval, the present invention adopts the firefly algorithm to set the corresponding error interval matrix for the generator equipment operation data matrix, thereby reducing the error in the setting process. At the same time, by simulating the behavior of fireflies being attracted by fireflies with high brightness, a mathematical model is constructed, which improves the speed and efficiency of the optimization process and effectively avoids the optimization results from falling into the local optimum.

[0066] Preferably, the specific steps of randomly selecting a generator set as a test device, obtaining a test device set, collecting the test taker's answer result data, and obtaining the test taker's answer result data matrix are as follows:

[0067] From the generator set A = {a 1 ,a 2 ,…,a i ,…,a k} randomly selects the generator set equipment as the test equipment, and obtains the test equipment set A'={a' 1 ,a' 2 ,…,a' i ,…,a' o}, where a' i represents the i-th generator set selected as the test equipment, and o represents the total number of generator sets selected as the test equipment;

[0068] The test question management center pushes the test question device set to the display screen used by the examinee to answer through the wireless communication network. The examinee answers according to the test question device set, collects the examinee's answer result data, and obtains the examinee's answer result data matrix D as follows:

[0069]

[0070] Among them, d ij It represents the answer result data given by the examinee to the jth type of operating parameters during the operation of the equipment in the i-th test question.

[0071] Generator set equipment is randomly selected as the test equipment, and candidates are asked to answer according to the test equipment. The data of candidates' answers are collected to ensure the reliability and comprehensiveness of the answer results.

[0072] Preferably, the specific steps of scoring the candidate's response results by using the error interval matrix and the candidate's response result data matrix are as follows:

[0073] Determine in sequence whether each candidate response result data in the candidate response result data matrix D is within the corresponding error interval in the error interval matrix Q, and count the total number λ of candidate response result data in the candidate response result data matrix D that is within the corresponding error interval in the error interval matrix Q;

[0074] Calculate the score of the candidate's response The candidate's response result data and the candidate's response result score are stored in the candidate data management module.

[0075] The candidates' responses are scored using the candidate response data matrix and error range matrix to avoid errors caused by manual review, reduce the consumption of human resources, and ensure the accuracy of the scoring results.

[0076] The present invention also discloses a system for an intelligent assessment method of a generator set monitor, the system comprising a candidate's facial feature image data acquisition module, a cheating behavior determination module, a generator set equipment operation data acquisition module, an error interval setting module, a candidate's answer result data collection module, and a candidate's answer result scoring module;

[0077] The candidate face feature image data acquisition module collects image data of the candidate during the answering process at irregular intervals through the image acquisition device, and performs low-pass filtering on the obtained image data to obtain the candidate face feature image data set;

[0078] The cheating behavior determination module sequentially matches each candidate's facial feature image data in the candidate's facial feature image data set with all the standard facial feature image data in the established standard facial feature image data set, calculates the candidate's facial image standard value according to the matching result, and compares the value with the set candidate's facial image standard threshold value, and determines whether the candidate has cheated according to the comparison result;

[0079] The generator set equipment operation data acquisition module acquires the operation data of the generator set equipment during the working process through the set generator set equipment set and the equipment operation parameter set to obtain the generator set equipment operation data matrix;

[0080] The error interval setting module sets the error interval of each generator set equipment operation data in the generator set equipment operation data matrix by using the firefly optimization algorithm to obtain an error interval matrix;

[0081] The candidate answer result data collection module randomly selects a generator set device from the generator set device set as the test equipment, and the candidate answers according to the test equipment set pushed by the test management center to the display screen used by the candidate to answer, collects the candidate answer result data, and obtains the candidate answer result data matrix;

[0082] The candidate response result scoring module calculates the score of the candidate response result according to the statistical results by counting the total number of candidate response result data in the candidate response result data matrix that are within the corresponding error interval.

[0083] (III) Beneficial effects

[0084] 1. The present invention obtains the facial feature image data of the examinee by pre-processing the image data collected by the image collection device at irregular intervals, thereby ensuring the randomness and universality of the collected image data; sequentially matches the facial feature image data of each examinee with all the standard facial feature image data in the standard facial feature image data set, and determines whether the examinee has cheated according to the matching result, thereby ensuring the reliability and accuracy of the determination result; randomly selects a generator set device as the test question device, and asks the examinee to answer according to the test question device, collects the answer result data of the examinee, and scores the examinee's answer result in combination with the error interval matrix, thereby avoiding errors caused by manual review, reducing the consumption of human resources, and ensuring the accuracy of the scoring result;

[0085] 2. Compared with the method of setting the error interval by calling historical data, the present invention adopts the firefly algorithm to set the corresponding error interval matrix for the generator equipment operation data matrix, which reduces the error in the setting process. At the same time, by simulating the behavior of fireflies being attracted by fireflies with high brightness, a mathematical model is constructed to improve the speed and efficiency of the optimization process, and effectively avoid the optimization result from falling into the local optimum;

[0086] 3. The present invention realizes image matching between the facial feature image data of the examinee and the standard facial feature image data through the beetle whisker algorithm, completes the accurate matching of the facial feature image, effectively eliminates the possibility of cheating, and realizes intelligent supervision during the examination. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0088] Figure 1 A flow chart of an intelligent assessment method for generator panel monitors provided by the present invention;

[0089] Figure 2 A module schematic diagram of a system for an intelligent assessment method for generator set monitoring personnel provided by the present invention. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0092] Embodiment 1 is as follows:

[0093] See also Figure 1 , an intelligent assessment method for generator panel monitors, tracking whether candidates have cheated during the assessment process and scoring the candidates' responses; including the following steps:

[0094] The image data of the examinee's answering process is collected and preprocessed to obtain the examinee's facial feature image data set; the steps include:

[0095] The image data of the examinee's answering process is collected from time to time by the image acquisition device installed on the display screen used by the examinee to answer, and the examinee's face image set E = {e 1 ,e 2 ,…,e i ,…,e p}, where e i represents the i-th examinee face image collected by the image collection device, and p represents the total number of examinee face images collected;

[0096] The sliding average filter is used to perform low-pass filtering on each candidate's face image in the candidate's face image set to obtain the candidate's face feature image data set E' = {e' 1 ,e' 2 ,…,e' i ,…,e' p}, where e' i It represents the facial feature image data of the examinee obtained after the i-th examinee's facial image is processed by low-pass filtering.

[0097] Establishing a standardized facial feature image data set, matching each candidate's facial feature image data in the candidate's facial feature image data set with each standardized facial feature image data in the standardized facial feature image data set in turn, and determining whether the candidate has cheated based on the matching results; comprising the following steps:

[0098] Establish a standard facial feature image dataset G = {g 1 ,g 2 ,…,g i ,…,g u}, where g i represents the i-th standard facial feature image data, and u represents the total number of standard facial feature image data;

[0099] Perform image matching on each candidate's facial feature image data in the candidate's facial feature image data set with all the standard facial feature image data in the standard facial feature image data set in sequence;

[0100] The image matching comprises the following steps:

[0101] Construct a population of longhorn beetles, set the population size of longhorn beetles to N', the second current number of iterations to T, and the second maximum number of iterations to T max And the dimension of the search space of normalized facial feature image data is M;

[0102] Each individual longhorn beetle in the longhorn beetle population corresponds to a standard facial feature image data in the standard facial feature image data set, and the initial position set Y of the longhorn beetle population is randomly generated in the standard facial feature image data search space. 1 ,Y 2 ,…,Y i ,…,Y p}, where Y i represents the initial position of the i-th individual in the beetle population;

[0103] Calculate the fitness function value of each individual beetle in the beetle population, and take the individual beetle with the highest fitness function value as the current optimal individual; the fitness function formula is as follows:

[0104]

[0105] Among them, F(y i ) represents the fitness function value of the i-th individual in the beetle population, y i represents the degree of deviation between the standard facial feature image data corresponding to the i-th longicorn individual and the facial feature image data of the candidate being matched, and φ represents the second correction value;

[0106] The iteration operation begins. During each iteration, each individual beetle in the beetle population uses the two antennae on the left and right sides of its head to detect the concentration of food odor in the standard facial feature image data search space, selects a direction to update its position according to the concentration of the food odor, recalculates the fitness function value of each individual beetle after the position update, and selects the individual beetle with the highest fitness function value as the current optimal individual.

[0107] Determine whether the second current iteration number T is less than the second maximum iteration number T max , if the second current iteration number T is less than the second maximum iteration number T max , then the second current iteration number is T plus 1, and the iteration operation is performed again; otherwise, the current optimal individual is taken as the global optimal solution, and the fitness function threshold is set. When the fitness function value of the global optimal solution is greater than the fitness function threshold, it is determined that the standard facial feature image data corresponding to the global optimal solution matches the facial feature image data of the candidate to be matched successfully; otherwise, it is determined that the facial feature image data of the candidate to be matched and all the standard facial feature image data in the standard facial feature image data set are not matched successfully;

[0108] Count the total number of candidate facial feature image data that successfully matches the standard facial feature image data in the candidate facial feature image dataset σ, and calculate the standard value of the candidate facial image

[0109] Determine whether the examinee has cheated; including the following steps:

[0110] Set the standard threshold of the examinee's face image to υ;

[0111] If ω≥υ, it is determined that the examinee has not cheated, and the examinee's facial feature image data is saved in the examinee data management module;

[0112] If ω<υ, it is determined that the examinee may have cheated, and the examinee's facial feature image data is fed back to the main control unit for secondary determination. After the secondary determination is completed, the examinee's facial feature image data is saved in the examinee data management module.

[0113] Setting a generator set equipment set and an equipment operation parameter category set, collecting operation data of the generator set equipment during operation according to the generator set equipment set and the equipment operation parameter category set, and obtaining a generator set equipment operation data matrix;

[0114] Set the generator set A = {a 1 ,a 2 ,…,a i ,…,a k}, where a irepresents the i-th generator set, k represents the total number of generator sets;

[0115] Set the equipment operation parameter category set B = {b 1 ,b 2 ,…,b i ,…,b l}, where b i represents the i-th type of operating parameter in the working process of the generator set equipment, and l represents the total number of operating parameter categories in the working process of the generator set equipment;

[0116] According to the generator set A={a 1 ,a 2 ,…,a i ,…,a k} and the equipment operation parameter category set B = {b 1 ,b 2 ,…,b i ,…,b l The operating data of the generator set equipment during operation is collected to obtain the generator set equipment operating data matrix C as follows:

[0117]

[0118] Among them, c ij Represents the data of the jth type of operating parameters during the operation of the i-th generator set.

[0119] The error interval of each generator set equipment operation data in the generator set equipment operation data matrix is ​​set by an intelligent optimization algorithm to obtain an error interval matrix; the process includes the following steps:

[0120] Construct a firefly population, set the firefly population size to N, the light intensity absorption coefficient to γ, the step size perturbation factor to α, the first current iteration number to t, and the first maximum iteration number to t max and the dimension of the running parameter search space is P';

[0121] Based on any one of the generator set operation data in the generator set operation data matrix C, N operation parameter intervals are randomly generated. Each firefly individual in the firefly population represents an operation parameter interval. The initial position set of the firefly population is set to X = {X 1 ,X 2 ,…,X i ,…,X N}, where X i represents the initial position of the i-th firefly individual in the firefly population;

[0122] Calculate the fitness function value of each firefly individual in the firefly population, and take the firefly individual with the highest fitness function value as the current optimal individual; the fitness function value formula is as follows:

[0123]

[0124] Among them, Fit(x i ) represents the fitness function value of the i-th firefly individual in the firefly population, x i represents the effect of the i-th firefly individual corresponding to the operating parameter interval on maintaining the normal operation of the generator set equipment, and ε represents the first correction value;

[0125] Start iterating and calculate the spatial distance between individual fireflies in the firefly population; the spatial distance calculation formula is as follows:

[0126]

[0127] Among them, D i,j represents the spatial distance between the i-th firefly individual and the j-th firefly individual, X i,z and X j,z They represent the positions of the i-th firefly individual and the j-th firefly individual in the z-th dimension operating parameter search space respectively;

[0128] In the operating parameter search space, firefly individuals with low fitness function values ​​are attracted to firefly individuals with high fitness function values; the attraction function formula is as follows:

[0129]

[0130] Among them, β(D i,j ) represents the attraction between the i-th firefly individual and the j-th firefly individual, β 0 It represents the attraction between two fireflies when the distance between them is 0;

[0131] Firefly individuals will be attracted by fireflies with high fitness function values ​​to update their positions in the operating parameter search space; the position update formula is as follows:

[0132]

[0133] in, and They represent the position of the i-th firefly individual in the firefly population during the t+1th and tth iterations, respectively. represents the position of the j-th firefly individual in the firefly population during the t-th iteration, and rand represents a random number between (0, 1);

[0134] Since the current best individual is the firefly individual with the highest fitness function value, the current best individual is set to randomly select a direction to update its position in the running parameter search space;

[0135] Calculate the fitness function value of each firefly individual in the firefly population after the position is updated, and take the firefly individual with the highest fitness function value as the current optimal individual;

[0136] Determine whether the first current iteration number t is less than the first maximum iteration number t max , if the first current iteration number t is less than the first maximum iteration number t max , then the first current iteration number t is increased by 1, and the iteration operation is performed again; otherwise, the current optimal individual is taken as the global optimal solution, and the operating parameter interval corresponding to the global optimal solution is output as the error interval of the operating data of the generator set equipment;

[0137] Repeat the above operation until the error intervals of all the generator set equipment operation data in the generator set equipment operation data matrix are obtained, and the error interval matrix Q is generated as follows:

[0138]

[0139] in, It represents the error interval of the jth type of operating parameters during the operation of the i-th generator set.

[0140] Randomly select a generator set as a test device to obtain a test device set, collect the test taker's answer result data, and obtain the test taker's answer result data matrix; including the following steps:

[0141] From the generator set A = {a 1 ,a 2 ,…,a i ,…,a k} randomly selects the generator set equipment as the test equipment, and obtains the test equipment set A'={a' 1 ,a' 2 ,…,a' i ,…,a' o}, where a' i represents the i-th generator set selected as the test equipment, and o represents the total number of generator sets selected as the test equipment;

[0142] The test question management center pushes the test question device set to the display screen used by the examinee to answer through the wireless communication network. The examinee answers according to the test question device set, collects the examinee's answer result data, and obtains the examinee's answer result data matrix D as follows:

[0143]

[0144] Among them, d ij It represents the answer result data given by the examinee to the jth type of operating parameters during the operation of the equipment in the i-th test question.

[0145] Scoring the candidate response results through the error interval matrix and the candidate response result data matrix; including the following steps:

[0146] Determine in sequence whether each candidate response result data in the candidate response result data matrix D is within the corresponding error interval in the error interval matrix Q, and count the total number λ of candidate response result data in the candidate response result data matrix D that is within the corresponding error interval in the error interval matrix Q;

[0147] Calculate the score of the candidate's response The candidate's response result data and the candidate's response result score are stored in the candidate data management module.

[0148] Embodiment 2 is as follows:

[0149] See also Figure 2 , a system for intelligent assessment of power generation unit monitoring personnel, including a candidate's facial feature image data acquisition module, a cheating behavior determination module, a power generation unit equipment operation data acquisition module, an error interval setting module, a candidate's answer result data collection module, and a candidate's answer result scoring module;

[0150] The candidate face feature image data acquisition module collects image data of the candidate during the answering process at irregular intervals through the image acquisition device, and performs low-pass filtering on the obtained image data to obtain the candidate face feature image data set;

[0151] The cheating behavior determination module matches each candidate's facial feature image data in the candidate's facial feature image data set with all the standard facial feature image data in the established standard facial feature image data set in turn, calculates the candidate's facial image standard value based on the matching result, and compares the value with the set candidate's facial image standard threshold value, and determines whether the candidate has cheated based on the comparison result.

[0152] The generator set equipment operation data acquisition module acquires the operation data of the generator set equipment during the working process through the set generator set equipment set and the equipment operation parameter set to obtain the generator set equipment operation data matrix;

[0153] The error interval setting module sets the error interval of each generator set equipment operation data in the generator set equipment operation data matrix by using the firefly optimization algorithm to obtain an error interval matrix;

[0154] The candidate answer result data collection module randomly selects a generator set device from the generator set device set as the test equipment, and the candidate answers according to the test equipment set pushed by the test management center to the display screen used by the candidate to answer, collects the candidate answer result data, and obtains the candidate answer result data matrix;

[0155] The candidate response result scoring module calculates the score of the candidate response result according to the statistical results by counting the total number of candidate response result data in the candidate response result data matrix that are within the corresponding error interval.

[0156] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0157] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. An intelligent assessment method for generator panel supervisors, characterized in that: Track whether candidates cheat during the assessment process and score the candidates' responses; The steps include: Collect image data during the examinee's answering process and perform preprocessing to obtain the examinee's facial feature image dataset; Establish a standardized facial feature image data set, match each candidate's facial feature image data in the candidate's facial feature image data set with each standardized facial feature image data in the standardized facial feature image data set in turn, and determine whether the candidate has cheated based on the matching results; Setting a generator set equipment set and an equipment operation parameter category set, collecting operation data of the generator set equipment during operation according to the generator set equipment set and the equipment operation parameter category set, and obtaining a generator set equipment operation data matrix; The error interval of each generator set equipment operation data in the generator set equipment operation data matrix is ​​set by an intelligent optimization algorithm to obtain an error interval matrix; Randomly select the generator set equipment as the test equipment to obtain the test equipment set, collect the test taker's answer result data, and obtain the test taker's answer result data matrix; The candidate response results are scored using the error interval matrix and the candidate response result data matrix.

2. The intelligent assessment method for generator monitoring personnel according to claim 1 is characterized in that: The collecting and preprocessing of image data during the examinee's answering process to obtain the examinee's facial feature image data set comprises the following steps: The image data of the examinee's answering process is collected from time to time by an image collection device installed on the display screen used by the examinee to obtain a set of examinee face images; A sliding average filter is used to perform low-pass filtering on each candidate's face image in the candidate's face image set to obtain a candidate's face feature image data set.

3. The intelligent assessment method for generator panel monitors according to claim 2 is characterized in that: The method of establishing a standard facial feature image data set, matching each candidate's facial feature image data in the candidate's facial feature image data set with each standard facial feature image data in the standard facial feature image data set in sequence, and determining whether the candidate has cheated according to the matching result comprises the following steps: Establish a standardized facial feature image dataset; Perform image matching on each candidate's facial feature image data in the candidate's facial feature image data set with all the standard facial feature image data in the standard facial feature image data set in sequence; Count the total number of candidate facial feature image data that successfully matches the standard facial feature image data in the candidate facial feature image dataset σ, and calculate the standard value of the candidate facial image Determine whether the examinee has engaged in cheating.

4. The intelligent assessment method for generator panel monitors according to claim 3 is characterized in that: The image matching comprises the following steps: Construct a population of longhorn beetles, set the population size of longhorn beetles to N', the second current number of iterations to T, and the second maximum number of iterations to T max And the dimension of the search space of normalized facial feature image data is M; Each individual longicorn beetle in the longicorn beetle population corresponds to a standard human face feature image data in the standard human face feature image data set, and an initial position set of the longicorn beetle population is randomly generated in the standard human face feature image data search space; Calculate the fitness function value of each individual beetle in the beetle population, and take the individual beetle with the highest fitness function value as the current optimal individual; The iteration operation begins. During each iteration, each individual beetle in the beetle population uses the two antennae on the left and right sides of its head to detect the concentration of food odor in the standard facial feature image data search space, selects a direction to update its position according to the concentration of the food odor, recalculates the fitness function value of each individual beetle after the position update, and selects the individual beetle with the highest fitness function value as the current optimal individual. Determine whether the second current iteration number T is less than the second maximum iteration number T max , if the second current iteration number T is less than the second maximum iteration number T max , then the second current iteration number is T plus 1, and the iteration operation is performed again; otherwise, the current optimal individual is taken as the global optimal solution, and the fitness function threshold is set. When the fitness function value of the global optimal solution is greater than the fitness function threshold, it is determined that the standard facial feature image data corresponding to the global optimal solution matches the facial feature image data of the candidate to be matched successfully; otherwise, it is determined that the facial feature image data of the candidate to be matched and all the standard facial feature image data in the standard facial feature image data set are not matched successfully.

5. The intelligent assessment method for generator panel supervisors according to claim 4 is characterized in that: The determination of whether the examinee has cheated includes the following steps: Set the standard threshold of the examinee's face image to υ; If ω≥υ, it is determined that the examinee has not cheated, and the examinee's facial feature image data is saved in the examinee data management module; If ω<υ, it is determined that the examinee may have cheated, and the examinee's facial feature image data is fed back to the main control unit for secondary determination. After the secondary determination is completed, the examinee's facial feature image data is saved in the examinee data management module.

6. The intelligent assessment method for generator panel monitors according to claim 5 is characterized in that: The setting of the generator set equipment set and the equipment operation parameter category set, and collecting the operation data of the generator set equipment during the working process according to the generator set equipment set and the equipment operation parameter category set comprises the following steps: Set the generator set equipment set and equipment operation parameter category set; The operating data of the generator set equipment during operation is collected according to the generator set equipment set and the equipment operating parameter set to obtain the generator set equipment operating data matrix C.

7. The intelligent assessment method for generator panel supervisors according to claim 6 is characterized in that: The error interval of each generator set equipment operation data in the generator set equipment operation data matrix is ​​set by the intelligent optimization algorithm to obtain the error interval matrix, which comprises the following steps: Construct a firefly population, set the firefly population size to N, the light intensity absorption coefficient to γ, the step size perturbation factor to α, the first current iteration number to t, and the first maximum iteration number to t max and the dimension of the running parameter search space is P'; Based on any one of the generator set equipment operation data in the generator set equipment operation data matrix C, N operation parameter intervals are randomly generated, and each firefly individual in the firefly population represents an operation parameter interval; the initial position of the firefly population is set; Calculate the fitness function value of each firefly individual in the firefly population, and take the firefly individual with the highest fitness function value as the current optimal individual; Start iterating and calculate the spatial distance between individual fireflies in the firefly population; In the operation parameter search space, firefly individuals with low fitness function values ​​are attracted to firefly individuals with high fitness function values; Firefly individuals will be attracted by fireflies with high fitness function values ​​to update their positions in the operating parameter search space; Since the current best individual is the firefly individual with the highest fitness function value, the current best individual is set to randomly select a direction to update its position in the running parameter search space; Calculate the fitness function value of each firefly individual in the firefly population after the position is updated, and take the firefly individual with the highest fitness function value as the current optimal individual; Determine whether the first current iteration number t is less than the first maximum iteration number t max , if the first current iteration number t is less than the first maximum iteration number t max , then the first current iteration number t is increased by 1 and the iteration operation is performed again; otherwise, the current optimal individual is taken as the global optimal solution, and the operating parameter interval corresponding to the global optimal solution is output as the error interval of the operating data of the generator set equipment; Repeat the above operation until the error intervals of all the generator set equipment operation data in the generator set equipment operation data matrix are obtained, and the error interval matrix Q is generated.

8. The intelligent assessment method for generator panel supervisors according to claim 7 is characterized in that: The randomly selecting a generator set device as a test device, obtaining a test device set, collecting the test taker's answer result data, and obtaining the test taker's answer result data matrix comprises the following steps: Randomly select a generator set device from the generator set device set as a test device to obtain a test device set; The test question management center pushes the test question device set to the display screen used by the examinee to answer via the wireless communication network. The examinee answers according to the test question device set, collects the examinee's answer result data, and obtains the examinee's answer result data matrix D.

9. The intelligent assessment method for generator panel monitors according to claim 8 is characterized in that: The step of scoring the candidate's response results by using the error interval matrix and the candidate's response result data matrix comprises the following steps: Determine in sequence whether each candidate response result data in the candidate response result data matrix D is within the corresponding error interval in the error interval matrix Q, and count the total number λ of candidate response result data in the candidate response result data matrix D that is within the corresponding error interval in the error interval matrix Q; Calculate the score of the candidate's response The candidate's response result data and the candidate's response result score are stored in the candidate data management module.

10. A system for implementing the intelligent assessment method for generator set monitoring personnel as described in any one of claims 1 to 9.

Images