CN114238719A - Simulation drill evaluation method - Google Patents

Simulation drill evaluation method Download PDF

Info

Publication number
CN114238719A
CN114238719A CN202111589731.0A CN202111589731A CN114238719A CN 114238719 A CN114238719 A CN 114238719A CN 202111589731 A CN202111589731 A CN 202111589731A CN 114238719 A CN114238719 A CN 114238719A
Authority
CN
China
Prior art keywords
data
evaluation
value
fusion
drilling
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111589731.0A
Other languages
Chinese (zh)
Inventor
罗小娅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sichuan Qiruike Technology Co Ltd
Original Assignee
Sichuan Qiruike Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sichuan Qiruike Technology Co Ltd filed Critical Sichuan Qiruike Technology Co Ltd
Priority to CN202111589731.0A priority Critical patent/CN114238719A/en
Publication of CN114238719A publication Critical patent/CN114238719A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/9035—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization

Abstract

The invention discloses a simulation drill evaluation method, which is based on the processing thought of multi-sensor fusion, each evaluator is abstracted into a sensor, the grade of each step of the evaluator is used as the measured value of the sensor at one moment, the weight of the evaluator is adjusted by measuring noise, and even if the evaluator is not a real sensor, the evaluator has the data attribute acquired by the sensor through abstraction. And combining a Kalman filtering algorithm, realizing the fusion of evaluation results of multiple evaluators, and solving the problems of discrete independence of evaluation time and evaluators.

Description

Simulation drill evaluation method
Technical Field
The invention relates to the technical field of data analysis, in particular to a simulation drilling evaluation method.
Background
Currently, a weighted summation mode is generally adopted for drill evaluation, namely, a score is set in each step, each evaluator sets a weight, and then a final result is obtained through weighted summation. The method disperses and separates each step, ignores the relevance among the steps, and also disperses the result of an evaluator, so that the time continuity of the evaluation of the drilling process cannot be highlighted.
The basic principle of multi-sensor information fusion is the process of comprehensively processing information by human brain, namely, a plurality of sensor resources are fully utilized, and the complementation and the redundancy information of various sensors in space and time are combined according to a certain optimization criterion through the reasonable control and use of various sensors and the observation information thereof, so that a more accurate and reliable conclusion is drawn.
Disclosure of Invention
The invention provides a simulation drilling evaluation method, which mainly solves the problems of drilling evaluation time and discrete independence of evaluators, and realizes the fusion of evaluation data of multiple evaluators through dynamic calculation instead of weighted summation; and (3) continuously updating the evaluation drilling time and personnel by referring to a multi-sensor fusion method and combining a Kalman filtering algorithm.
The invention achieves the above object by the following technical inventions:
a simulation drill evaluation method comprises the following steps:
step 1, construction of drilling evaluation data: the evaluation personnel evaluate the conditions of each step of the drilling personnel, and the evaluation results are written into the database;
step 2, inquiring initial step data: querying data of all data sources in the database with the step 1, namely, all evaluation personnel evaluate the step 1;
step 3, establishing an initial value: selecting the data of the first data source inquired in the step 2 to establish an initial value, wherein the initial value comprises an evaluated value and the drilling data noise corresponding to the data source;
step 4, traversing the query data in the step 2, receiving the residual unprocessed data in the step 3, and performing data correction operation on the data;
step 5, writing the fusion data generated in the step 4 into a database;
step 6, performing data prediction operation on the fusion data in the step 4 to predict the state of the next process;
step 7, inquiring drill evaluation data of all data sources of the next process;
step 8, traversing all the drilling evaluation data inquired in the step 7, correcting the state in the step 6, and executing data correction operation;
step 9, writing the fusion data generated in the step 8 into a database;
step 10 repeats steps 6 through 9 until the drill phase is complete.
The further invention is that, in the step 1, two aspects are mainly included:
1-1, data source: an evaluator is a data source, and the id of the data source is the unique identification information of the evaluator, such as data source A, B;
1-2, data value: each evaluator evaluates each step in the drill process as a data value, and each data value is bound with a data source and a step number, such as V { data source: a, step number: 1, value: 95}.
Further, in the step 3, the data source noise is constructed by: the noise of the drilling data is used for adjusting the weight of each data source (evaluator), namely the observation noise of a Kalman filter, and the noise generally reflects the accuracy of an observation value (evaluation result of the evaluator in the invention) and an actual situation, in the invention, the noise range is (0.1-0.9), and for the evaluator with higher weight value, the noise of the drilling data is set to be smaller and can approach to 0.1, namely the evaluation value is considered to be more credible, and the gradient of the fusion result towards the evaluation value is larger; on the contrary, for the evaluators with lower weight values, the noise setting of the drilling data is larger and can approach to 0.9, and the noise value can be adjusted according to the actual drilling situation and different evaluators, and the adjusting range is (0.1-0.9).
Further, in the step 5, the global fusion object is constructed by: the global fusion object is a certain data source value and corresponding data source noise of the initial step initially, and the subsequent and input drilling evaluation data are fused according to a fusion method, so that the fusion evaluation data of each step are output.
The invention has the beneficial effects that:
the simulation drill evaluation method is based on a processing thought of multi-sensor fusion, each evaluator is abstracted into a sensor, the grade of each step of the evaluator is used as a measured value of the sensor at one moment, and the weight of the evaluator is adjusted through measuring noise. So, even if the rater is not a real sensor, by abstraction, the rater's family personnel are provided with the data attributes collected by the sensor. And combining a Kalman filtering algorithm, realizing the fusion of evaluation results of multiple evaluators, and solving the problems of discrete independence of evaluation time and evaluators.
Drawings
In order to more clearly illustrate the technical invention in the embodiments of the present invention, the drawings which are needed to be practical in the embodiments or the prior art description will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without inventive work.
FIG. 1 is a diagram of an overall technical invention architecture.
FIG. 2 is a flow chart of the present invention.
Detailed Description
The technical invention of the present invention will be described in detail below in order to make the objects, technical inventions and advantages of the present invention more apparent. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without any inventive step, are within the scope of the present invention.
In any embodiment, as shown in fig. 1-2, a simulation drill evaluation method of the present invention includes:
step 1, construction of drilling evaluation data: the evaluation personnel evaluate the conditions of each step of the drilling personnel, and the evaluation results are written into the database;
step 2, inquiring initial step data: querying data of all data sources in the database with the step 1, namely, all evaluation personnel evaluate the step 1;
step 3, establishing an initial value: selecting the data of the first data source inquired in the step 2 to establish an initial value, wherein the initial value comprises an evaluated value and the drilling data noise corresponding to the data source;
step 4, traversing the query data in the step 2, receiving the residual unprocessed data in the step 3, and performing data correction operation on the data;
step 5, writing the fusion data generated in the step 4 into a database;
step 6, performing data prediction operation on the fusion data in the step 4 to predict the state of the next process;
step 7, inquiring drill evaluation data of all data sources of the next process;
step 8, traversing all the drilling evaluation data inquired in the step 7, correcting the state in the step 6, and executing data correction operation;
step 9, writing the fusion data generated in the step 8 into a database;
step 10 repeats steps 6 through 9 until the drill phase is complete.
And (3) construction of drilling evaluation data: the drill evaluation data is evaluated by an evaluator, and the evaluation result is stored in a database, and the method mainly comprises the following two aspects:
1-1, data source: an evaluator is a data source, and the id of the data source is the unique identification information of the evaluator, such as data source A, B;
1-2, data value: each evaluator evaluates each step in the drill process as a data value, and each data value is bound with a data source and a step number, such as V { data source: a, step number: 1, value: 95}.
Data source noise construction: the noise of the drilling data is used for adjusting the weight of each data source (evaluator), namely the observation noise of a Kalman filter, and the noise generally reflects the accuracy of an observation value (evaluation result of the evaluator in the invention) and an actual situation, in the invention, the noise range is (0.1-0.9), and for the evaluator with higher weight value, the noise of the drilling data is set to be smaller and can approach to 0.1, namely the evaluation value is considered to be more credible, and the gradient of the fusion result towards the evaluation value is larger; on the contrary, for the evaluators with lower weight values, the noise setting of the drilling data is larger and can approach to 0.9, and the noise value can be adjusted according to the actual drilling situation and different evaluators, and the adjusting range is (0.1-0.9).
Constructing a fusion object: the global fusion object is a certain data source value and corresponding data source noise of the initial step initially, and the subsequent and input drilling evaluation data are fused according to a fusion method, so that the fusion evaluation data of each step are output.
And (3) performing drilling evaluation data fusion: the method mainly comprises the following steps:
1. data synchronization: the method has the main functions of synchronizing the step numbers of different data sources and guaranteeing that the evaluation result of the same step is corrected during fusion correction.
2. And (3) data filtering: the method has the main function of filtering abnormal evaluation values, and the filtering rule of the abnormal values is preset according to the actual situation.
3. And (3) data prediction: and when a fusion condition is triggered (the fusion calculation in the current step is completed, and the fusion calculation in the next step is carried out), predicting the value of the global fusion object according to the Kalman filter state transition matrix, and predicting the evaluation result in the next step.
Figure BDA0003428701440000051
Figure BDA0003428701440000052
Figure BDA0003428701440000053
Represents the estimated value of the prior state at the time k, which is an unreliable estimate made by the algorithm based on the result of the previous iteration (i.e., the posterior estimated value of the last cycle).
Figure BDA0003428701440000054
And the posterior state estimated value at the k-1 moment is represented, namely the optimal estimated value at the moment to be output, and the value is the result of Kalman filtering.
A: the state transition matrix is represented by an n × n order square matrix, which is the basis for predicting the state variables by the algorithm, and if the state transition matrix does not accord with the target model, filtering divergence can be caused.
B: represents the gain of the optional control input u e Rl, which in the present invention is 0.
uk-1: the control gain at time k-1 is shown and is set to 0 in the present invention.
Figure BDA0003428701440000061
The prior estimation covariance of k time is shown, the covariance matrix can be recurred later as long as the initial P0 is determined, and the initial covariance matrix P0 has little influence on the filtering effect and can be converged quickly as long as the initial covariance matrix P0 is not 0.
Figure BDA0003428701440000062
The a posteriori estimated covariance, representing the time instant k-1, is one of the filtering results.
Q: represents the covariance of the process excitation noise, which is the error between the state transition matrix and the actual process. The invention assumes that it is a fixed matrix and achieves better performance of the filter by finding the optimal Q value.
4. And (3) data correction: inquiring and acquiring the drilling evaluation data of all data sources of the corresponding steps, traversing the evaluation data, and repeatedly obtaining the drilling evaluation data of all data sources according to the evaluation data zkCorrecting the predicted result, and obtaining the result once every correction
Figure BDA0003428701440000063
The final result is obtained after traversing
Figure BDA0003428701440000064
Figure BDA0003428701440000065
Figure BDA0003428701440000066
Figure BDA0003428701440000067
Figure BDA0003428701440000068
And the posterior state estimated value at the k moment is represented, namely the optimal estimated value at the moment to be output, and the value is the result of Kalman filtering.
Figure BDA0003428701440000069
The a posteriori estimated covariance, representing time k, is one of the filtering results.
Kk: representing the kalman gain, is an intermediate result of the filtering.
zk: representing the measurement, is an m-th order vector.
H: the measurement matrix is represented as an m x n order matrix that converts m-dimensional measurements to n-dimensional values corresponding to state variables.
R: represents the measurement noise covariance, which is a numerical value that is related to the evaluator's weight, i.e., the data source noise.
5. Fusing result data output in each step: and writing the fusion data result of each step into a database, so that the fusion evaluation result of the drilling personnel at each step can be known by querying the database.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims. It should be noted that the various technical features described in the above embodiments can be combined in any suitable manner without contradiction, and the invention is not described in any way for the possible combinations in order to avoid unnecessary repetition. In addition, any combination of the various embodiments of the present invention is also possible, and the same should be considered as the disclosure of the present invention as long as it does not depart from the spirit of the present invention.

Claims (5)

1. A simulation drill evaluation method is characterized by comprising the following steps:
step 1, construction of drilling evaluation data: the evaluation personnel evaluate the conditions of each step of the drilling personnel, and the evaluation results are written into the database;
step 2, inquiring initial step data: querying data of all data sources in the database with the step 1, namely, all evaluation personnel evaluate the step 1;
step 3, establishing an initial value: selecting the data of the first data source inquired in the step 2 to establish an initial value, wherein the initial value comprises an evaluated value and the drilling data noise corresponding to the data source;
step 4, traversing the query data in the step 2, receiving the residual unprocessed data in the step 3, and performing data correction operation on the data;
step 5, writing the fusion data generated in the step 4 into a database;
step 6, performing data prediction operation on the fusion data in the step 4 to predict the state of the next process;
step 7, inquiring drill evaluation data of all data sources of the next process;
step 8, traversing all the drilling evaluation data inquired in the step 7, correcting the state in the step 6, and executing data correction operation;
step 9, writing the fusion data generated in the step 8 into a database;
step 10 repeats steps 6 through 9 until the drill phase is complete.
2. The simulation drill evaluation method of claim 1, wherein the step 1 mainly comprises two steps:
a data source: one evaluator is a data source, and the id of the data source is the unique identification information of the evaluator;
data value: and each evaluator evaluates each step in the drilling process as a data value, and each data value is bound with a data source and a step number.
3. The simulation drill evaluation method of claim 1, wherein in the step 3, the data source noise is constructed by: the noise of the drilling data is used for adjusting the weight of each data source, the noise value range is (0.1-0.9), the noise of the drilling data is set to be smaller and approaches to 0.1 for an evaluator with a higher weight value, namely the evaluation value is considered to be more credible, and the gradient of the fusion result towards the evaluation value is larger; on the contrary, for the evaluators with lower weight values, the noise setting of the drilling data is larger and approaches to 0.9, and the adjustment range is (0.1-0.9) according to the actual drilling situation and different evaluators.
4. The simulation drill evaluation method according to claim 1, wherein in the step 5, the global fusion object is constructed by: the global fusion object is a certain data source value and corresponding data source noise of the initial step initially, and the subsequent and input drilling evaluation data are fused according to a fusion method, so that the fusion evaluation data of each step are output.
5. The simulation drill evaluation method according to claim 1, wherein in the step 8, drill evaluation data are fused: the method mainly comprises the following steps:
a. data synchronization: synchronizing the step numbers of different data sources to ensure that the evaluation result of the same step is corrected when fusion correction is carried out;
b. and (3) data filtering: filtering the abnormal evaluation value, wherein the filtering rule of the abnormal value is preset according to the actual situation;
c. and (3) data prediction: when a fusion condition is triggered, predicting the value of the global fusion object according to a Kalman filter state transition matrix, and predicting the evaluation result of the next step;
Figure FDA0003428701430000021
Figure FDA0003428701430000022
Figure FDA0003428701430000023
representing the estimated value of the prior state at the moment k, which is unreliable estimation made by the algorithm according to the result of the previous iteration;
Figure FDA0003428701430000024
the posterior state estimated value at the k-1 moment is represented, namely the optimal estimated value at the moment to be output, and the value is the result of Kalman filtering;
a: the state transition matrix is represented as an n multiplied by n order square matrix which is the basis for predicting the state variable by the algorithm, and if the state transition matrix does not accord with the target model, filtering divergence can be caused;
b: represents the gain of the optional control input u e Rl, which is 0 in this step;
uk-1: represents the control gain at time k-1, and is set to 0 in this step;
Figure FDA0003428701430000025
representing the prior estimated covariance at time k;
Figure FDA0003428701430000031
representing the posteriori estimated covariance at time k-1;
q: covariance representing process excitation noise, which is the error between the state transition matrix and the actual process;
d. and (3) data correction: inquiring and obtaining the exercise evaluation number of all data sources of the corresponding step of predictionAccording to the evaluation data z, traversing the evaluation data and repeating the evaluation datakCorrecting the predicted result, and obtaining the result once every correction
Figure FDA0003428701430000032
The final result is obtained after traversing
Figure FDA0003428701430000033
Figure FDA0003428701430000034
Figure FDA0003428701430000035
Figure FDA0003428701430000036
Figure FDA0003428701430000037
The posterior state estimated value at the k moment is represented, namely the optimal estimated value at the moment to be output, and the value is the result of Kalman filtering;
Figure FDA0003428701430000038
representing the posterior estimated covariance at time k;
Kk: representing the kalman gain, which is an intermediate result of the filtering;
zk: representing the measured value, is an m-order vector;
h: the expression measurement matrix is an m multiplied by n order matrix which converts m dimension measurement values into n dimensions corresponding to the state variables;
r: represents the measurement noise covariance, which is a numerical value that is related to the evaluator weight, i.e., the data source noise;
e. fusing result data output in each step: and writing the fusion data result of each step into a database, and inquiring the database to know the fusion evaluation result of the drilling personnel at each step.
CN202111589731.0A 2021-12-23 2021-12-23 Simulation drill evaluation method Pending CN114238719A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111589731.0A CN114238719A (en) 2021-12-23 2021-12-23 Simulation drill evaluation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111589731.0A CN114238719A (en) 2021-12-23 2021-12-23 Simulation drill evaluation method

Publications (1)

Publication Number Publication Date
CN114238719A true CN114238719A (en) 2022-03-25

Family

ID=80762003

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111589731.0A Pending CN114238719A (en) 2021-12-23 2021-12-23 Simulation drill evaluation method

Country Status (1)

Country Link
CN (1) CN114238719A (en)

Similar Documents

Publication Publication Date Title
JP6384065B2 (en) Information processing apparatus, learning method, and program
JP6243080B1 (en) Preprocessor and abnormal sign diagnosis system
WO2019201081A1 (en) Method, device, and system for estimating causality between observation variables
US8849737B1 (en) Prediction method of predicting a future state of a system
CN109447152B (en) Fault prediction method based on Monte Carlo tree search and neural network
AU2020100630A4 (en) System and method for grade estimation using gradient boosted decesion tree based machine learning algorithims
CN113239022B (en) Method and device for complementing missing data in medical diagnosis, electronic device and medium
CN113037577A (en) Network traffic prediction method, device and computer readable storage medium
WO2017095590A1 (en) Stochastic map generation and bayesian update based on stereo vision
CN108764348A (en) Collecting method based on multiple data sources and system
Palm Multiple-step-ahead prediction in control systems with Gaussian process models and TS-fuzzy models
CN114238719A (en) Simulation drill evaluation method
WO2019012740A1 (en) Operation rule extraction device, operation rule extraction system, and operation rule extraction method
Aghapour et al. Performance specified state estimation with minimum risk
JP2021125136A (en) Optimization device and optimization method
Lu et al. Parameter estimation of software reliability growth models by a modified whale optimization algorithm
CN114513374B (en) Network security threat identification method and system based on artificial intelligence
Górak et al. Malfunction immune Wi–Fi localisation method
CN116166642A (en) Spatio-temporal data filling method, system, equipment and medium based on guide information
US11475255B2 (en) Method for adaptive context length control for on-line edge learning
JP6950647B2 (en) Data determination device, method, and program
KR101820635B1 (en) The method of knowledge diagnosis using knowledge map based on ontology
CN108960406B (en) MEMS gyroscope random error prediction method based on BFO wavelet neural network
KR20180020334A (en) Apparatus for forecasting disruption of communication system and that of method
CN116661426B (en) Abnormal AI diagnosis method and system of sensor operation control system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination