CN113505925A - ANFIS-based laboratory dangerous chemical abnormal information early warning method - Google Patents
ANFIS-based laboratory dangerous chemical abnormal information early warning method Download PDFInfo
- Publication number
- CN113505925A CN113505925A CN202110781857.1A CN202110781857A CN113505925A CN 113505925 A CN113505925 A CN 113505925A CN 202110781857 A CN202110781857 A CN 202110781857A CN 113505925 A CN113505925 A CN 113505925A
- Authority
- CN
- China
- Prior art keywords
- early warning
- anfis
- probability
- predicted
- output
- 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.)
- Granted
Links
- 239000000126 substance Substances 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 20
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 239000000383 hazardous chemical Substances 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000005284 excitation Effects 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 abstract description 7
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 239000003153 chemical reaction reagent Substances 0.000 description 3
- 230000006378 damage Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 231100000086 high toxicity Toxicity 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/043—Architecture, e.g. interconnection topology based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Tourism & Hospitality (AREA)
- Molecular Biology (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Fuzzy Systems (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Pure & Applied Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Automation & Control Theory (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a laboratory dangerous chemical abnormal information early warning method based on ANFIS, which belongs to the field of artificial intelligence and comprises a dose-time sequence early warning model and an ANFIS early warning model; the dose-time sequence early warning model is used for processing the predicted return time and the predicted used dose data in the laboratory information management system and finally outputting a probability P1 of whether early warning is performed or not; the ANFIS early warning model is composed of a five-layer ANFIS network structure and is used for processing the predicted return time and the predicted dosage data in the information management system after being trained through a data set, and finally outputting a probability P2 of whether early warning is performed or not; and finally, performing weighted calculation on the probabilities P1 and P2 obtained by the dose-time sequence early warning model and the ANFIS early warning model to obtain the final early warning probability P, and judging whether early warning is needed or not.
Description
Technical Field
The invention belongs to the field of artificial intelligence, and relates to an ANFIS-based laboratory dangerous chemical abnormal information early warning method.
Background
At present, most colleges and universities still have low-efficiency manpower management modes for the management of dangerous chemicals in laboratories, mainly adopt behind-mode hand account management or management through Excel, occupy labor cost, and because the management cost is high, the management of accurate dangerous chemical reagents, especially the management of abnormal use conditions of the reagents is complicated, and the use of dangerous chemical reagents cannot be accurate and effective.
Moreover, the college laboratory is an important place for teachers and students in colleges and universities to carry out experimental teaching and scientific research, and along with the gradual increase of the number of colleges and universities, the supervision perfection degree and the management practicability for the abnormal use condition of dangerous chemicals in the college laboratory are further enhanced. Most dangerous chemicals have the characteristics of flammability, explosiveness, high corrosiveness, high toxicity and the like, and are easy to cause dangers such as equipment damage, personnel injury and the like when the chemicals are used on the premise of not strict management.
Disclosure of Invention
In view of the above, the present invention provides an ANFIS-based laboratory hazardous chemical anomaly information early warning method, which solves the problem that in the conventional laboratory hazardous chemical management of colleges and universities, hazardous chemical usage anomaly information cannot be processed in time, resulting in irreparable loss.
In order to achieve the purpose, the invention provides the following technical scheme:
an ANFIS-based laboratory hazardous chemical anomaly information early warning method comprises the following steps:
s1: acquiring a laboratory hazardous chemical calling record containing the predicted return time and the predicted used dosage data;
s2: inputting the predicted return time and the predicted used dose data in the laboratory dangerous chemical calling record into a dose-time sequence early warning model to obtain an early warning probability P1;
s3: inputting the predicted return time and the predicted dosage data in the laboratory dangerous chemical calling record into an ANFIS early warning model to obtain an early warning probability P2;
s4: performing weighted calculation on the early warning probabilities P1 and P2 to obtain a final early warning probability P;
s5: and judging whether the early warning probability P exceeds an early warning threshold value, if not, giving no early warning, and if so, sending an early warning to an administrator.
Further, the laboratory hazardous chemical call record in step S1 is obtained from the laboratory information management system.
Further, the specific steps of the dose-time sequence early warning model outputting the early warning probability P1 in step S2 are as follows:
s21: inputting the predicted return time T1 and the predicted dose M1 in the laboratory hazardous chemical call record;
s22: setting threshold values T0 and M0, and respectively judging the predicted return time T1 and the predicted used dose M1;
s23: obtaining an early warning probability Pt according to an algebraic relation between the predicted return time T1 and a threshold T0, and obtaining an early warning probability Pm according to an algebraic relation between the predicted used dose M1 and a threshold M0;
s24: and performing weighted calculation on the early warning probabilities Pt and Pm to obtain the probability P1 of the dose-time sequence early warning model.
Further, in step S23, the algebraic relationship between the expected return time T1 and the threshold T0 is:
when T1 is less than or equal to T0, the output probability Pt is 0;
when T1 is more than T0 and T1 is less than or equal to 1.2 times of T0, the output probability Pt is 0.2;
when T1 is more than 1.2 times of T0 and T1 is less than or equal to 1.4 times of T0, the output probability Pt is 0.4;
when T1 is more than 1.4 times T0 and T1 is less than or equal to 1.6 times T0, the output probability Pt is 0.6;
when T1 is more than 1.6 times of T0 and T1 is less than or equal to 1.8 times of T0, the output probability Pt is 0.8;
when T1 is more than 1.8 times T0, the output probability Pt is 1.0;
the algebraic relationship between the expected dose M1 and the threshold M0 is:
when M1 is less than or equal to M0, the output probability Pm is 0;
when M1 is more than M0 and M1 is less than or equal to 1.2 times of M0, the output probability Pm is 0.2;
when M1 is more than 1.2 times of M0 and M1 is less than or equal to 1.4 times of M0, the output probability Pm is 0.4;
when M1 is more than 1.4 times of M0 and M1 is less than or equal to 1.6 times of M0, the output probability Pm is 0.6;
when M1 is more than 1.6 times of M0 and M1 is less than or equal to 1.8 times of M0, the output probability Pm is 0.8;
when M1 is more than 1.8 times M0, the output probability Pm is 1.0.
Further, the weighting calculation formula in step S24 is:
P1＝λ1Pt+(1-λ1)Pm
λ1a weight value in the range of 0-1.
Further, the ANFIS early warning model in step S3 adopts a five-layer fuzzy neural network structure, which is a fuzzy layer, a rule layer, a normalization layer, a de-fuzzy layer, and an output layer;
all nodes of the fuzzy layer are self-adaptive and are used for fuzzifying all input data according to a membership function, the adopted membership function is a triangular membership function, and the formula is as follows:
where { a, b, c } is a set of preconditions whose values are continuously updated during the training phase, parameters a and c determining the "feet" of the triangle, and parameter b determining the "peaks" of the triangle;
the output function expression of the fuzzification layer is as follows:
OB,j＝μBj(x2)，j＝1,2,3
OA,i、OB,jare respectively data x1，x2At the output of the blurring layer, wherein
And
are respectively data x1，x2A membership function of;
the output of the rule layer multiplies all input data to calculate the excitation intensity w of each rulekThe expression is as follows:
wk＝μA,i(x1)μB,j(x2)，i＝1,2,3,j＝1,2,3
the normalization layer normalizes the excitation intensity of the rule layer, and the calculation formula is as follows:
the de-fuzzy layer is used for calculating a weight value of each node, and the calculation formula is as follows:
wherein p isk、qkAnd kkIs a back-part parameter of the node;
the output layer combines the output of the de-fuzzy layer as the total output, and the calculation formula is as follows:
the output y value is the early warning probability P2 of the ANFIS early warning model.
Further, before executing step S3, the method further includes:
acquiring a laboratory hazardous chemicals call record containing predicted return time and predicted usage dosage data for training of an ANFIS network;
training an ANFIS network according to the laboratory dangerous chemical calling record for training to obtain an ANFIS early warning model to be tested;
acquiring a laboratory dangerous chemical calling record containing the predicted return time and the predicted using dosage data for testing the ANFIS network, and acquiring a test result of early warning of abnormal information of the laboratory dangerous chemical by the ANFIS network;
and reading the early warning result of the ANFIS early warning model to the data, comparing training data to obtain the early warning accuracy of the ANFIS early warning model, and taking the ANFIS network as a final ANFIS early warning model.
Further, the weighting calculation formula in step S4 is:
P＝λP1+(1-λ)P2
lambda is a weight value with a value range of 0-1.
The invention has the beneficial effects that: the invention solves the problem that the abnormal information of the dangerous chemical use in the dangerous chemical management of the laboratory can not be processed in time, which causes irreparable loss.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a schematic flow chart of an ANFIS-based laboratory hazardous chemical anomaly information early warning method according to the present invention;
FIG. 2 is a flow chart of a dose-timing early warning model of the present invention;
FIG. 3 is a diagram of a neural network architecture for the ANFIS early warning model of the present invention;
FIG. 4 is a flow chart of the probability processing of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "front", "rear", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not an indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes, and are not to be construed as limiting the present invention, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.
Referring to fig. 1 to 4, a method for early warning abnormal information of hazardous chemicals in a college laboratory is disclosed, where the method for early warning abnormal information of hazardous chemicals in a college laboratory includes: dose-timing early warning model, ANFIS early warning model; in a college laboratory information management system, determining a target chemical according to a calling record input in the system when a user calls the chemical, and sending the calling record to a dose-time sequence early warning model;
the early warning system for the abnormal information of the dangerous chemicals in the colleges and universities is characterized in that a calling record of the dangerous chemicals is input into the system, the system respectively and independently analyzes the predicted return time and the predicted used dose in the calling record through a dose-time sequence early warning model and an ANFIS early warning model to respectively obtain abnormal early warning probabilities, and finally, a final probability is obtained through weighting calculation to judge whether the dangerous chemicals need early warning;
the information contained in the chemical call record is: chemical information, time taken out, expected return time, use, expected dose, actual dose, applicant name, applicant's department;
referring to fig. 2, fig. 2 is a flow chart of a dose-timing model of the present invention. The working flow of the dose-time sequence model comprises the steps of firstly inputting predicted return time T1 and predicted used dose M1 in a dangerous chemical calling record, judging T1 and M1 according to a set return time threshold T0 and a set used dose threshold M0, and judging whether the probability of early warning is needed or not according to the interval of T1 and M1; when T1 ≦ T0, output probability Pt is 0, when T1 > T0 and T1 ≦ 1.2 times T0, output probability Pt is 0.2, when T1 > 1.2 times T0 and T1 ≦ 1.4 times T0, output probability Pt is 0.4, when T1 > 1.4 times T0 and T1 ≦ 1.6 times T0, output probability Pt is 0.6, when T1 > 1.6 times T0 and T1 ≦ 1.8 times T0, output probability Pt is 0.8, when T1 > 1.8 times T0, output probability Pt is 1.0; similarly, the same is true for the early warning probability relating to the dose used, where M1 ≦ M0, the output probability Pm is 0, when M1 > M0 and M1 ≦ 1.2 times M0, 0.2, when M1 > 1.2 times M0 and M1 ≦ 1.4 times M0, 0.4, when M1 > 1.4 times M0 and M1 ≦ 1.6 times M0, 0.6, when M1 > 1.6 times M0 and M1 ≦ 1.8 times M0, 0.8, and 1.0 when M1 > 1.8 times M0. According to the dosage-timeThe probabilities Pt and Pm obtained by the sequence early warning model adopt P1 ═ lambda1Pt+(1-λ1) The equation of Pm calculates the final probability of the dose-timing early warning model, where λ1Take 0.5.
Before the input predicted return time and predicted use dosage data are processed, the ANFIS early warning model needs to obtain laboratory dangerous chemical call records containing the predicted return time and the predicted use dosage data for training the ANFIS network, obtain the ANFIS early warning model to be tested after training, obtain another group of laboratory dangerous chemical call records containing the predicted return time and the predicted use dosage data for testing the ANFIS network again, and obtain the test result of the ANFIS early warning model for early warning the abnormal information of the laboratory dangerous chemicals; and comparing the test result with the training data to obtain the accuracy of the ANFIS early warning model, and if the accuracy is high enough, using the ANFIS network as the preset ANFIS early warning model.
The ANFIS network is a system model which is not deeply known by the fuzzy algorithm through combining a neural network and the fuzzy algorithm, namely, the ANFIS network can play a role in researching a control model of a system in depth due to objective reasons; the neural network is used for classification or regression prediction since being proposed, and the neural network can intelligently judge subsequent data through training and verification of a data set;
the ANFIS early warning model is characterized in that a sample data training set required to be used is derived from past data in a laboratory information management system of colleges and universities. The advantage of using the previous data is that the ANFIS model can be trained according to the accurate use of dangerous chemicals, the use time and the use amount each time, so that better judgment capability is obtained.
The model adopts an ANFIS network, data before training can be predicted and processed, and the data after training can be predicted and processed; the ANFIS early warning model comprises 5 layers in total, and is specifically divided into:
the first layer is a fuzzy layer, all nodes in the layer are self-adaptive and are used for fuzzifying all input data according to a membership function, the membership function adopted by the layer is a triangular membership function, and the formula is as follows:
The output function expression of the first layer is:
OB,j＝μBj(x2)，j＝1,2,3
OA,i、OB,jare each x1，x2An output of the first layer, wherein
And
are each x1，x2Membership function of (c).
The second layer is a rule layer, the output of which multiplies all the input data to calculate the excitation intensity w of each rulekThe expression is as follows:
wk＝μA,i(x1)μB,j(x2)，i＝1,2,3,j＝1,2,3
the input data are predicted return time T1 and predicted used dose M1 in the chemical calling record, and the "S", "M" and "L" represent small "," medium "and" large "in the fuzzy inference through the fuzzification processing, which is determined according to threshold values T0 and M0, because there are two input parameters and each parameter has three states of" S "," M "and" L ", there are 3 × 3 ═ 9 fuzzy rules in the fuzzy rule layer, namely, for O in the formula hereA,iAnd OB,jRespectively taking three values；
The third layer is a normalization layer, the excitation intensity of the second layer is normalized, and the calculation formula is as follows:
the fourth layer is a de-fuzzy layer, the weight value of each node is calculated, and the calculation formula is as follows:
The fifth layer is an output layer, the outputs of the de-fuzzy layers are merged and used as the total output, and the calculation formula is as follows:
the output y value is the early warning probability P2 of the ANFIS early warning model;
referring to fig. 4, fig. 4 is a flow chart of the probability processing of the present invention. After obtaining the probability P1 of the output of the dose-time sequence model and the probability P2 of the output of the ANFIS model, processing the two probabilities, and calculating the final probability P of the whole system by adopting the formula of P ═ lambda P1+ (1-lambda) P2, wherein the value of lambda is 0.5, and finally obtaining P; when P is less than or equal to 0.5, no early warning is carried out, when P is more than 0.5, the system carries out abnormal information early warning processing on the dangerous chemical calling record, namely, an early warning alarm is sent to an administrator, the administrator is required to contact the caller of the dangerous chemical calling record in the colleges and universities, and a warning is sent.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (8)
1. An ANFIS-based laboratory dangerous chemical abnormal information early warning method is characterized by comprising the following steps: the method comprises the following steps:
s1: acquiring a laboratory hazardous chemical calling record containing the predicted return time and the predicted used dosage data;
s2: inputting the predicted return time and the predicted used dose data in the laboratory dangerous chemical calling record into a dose-time sequence early warning model to obtain an early warning probability P1;
s3: inputting the predicted return time and the predicted dosage data in the laboratory dangerous chemical calling record into an ANFIS early warning model to obtain an early warning probability P2;
s4: performing weighted calculation on the early warning probabilities P1 and P2 to obtain a final early warning probability P;
s5: and judging whether the early warning probability P exceeds an early warning threshold value, if not, giving no early warning, and if so, sending an early warning to an administrator.
2. The ANFIS-based laboratory hazardous chemical anomaly information early warning method as claimed in claim 1, wherein: and the calling record of the dangerous chemicals in the laboratory in the step S1 is acquired from the laboratory information management system.
3. The ANFIS-based laboratory hazardous chemical anomaly information early warning method as claimed in claim 1, wherein: the specific steps of the dose-time sequence early warning model in the step S2 for outputting the early warning probability P1 are as follows:
s21: inputting the predicted return time T1 and the predicted dose M1 in the laboratory hazardous chemical call record;
s22: setting threshold values T0 and M0, and respectively judging the predicted return time T1 and the predicted used dose M1;
s23: obtaining an early warning probability Pt according to an algebraic relation between the predicted return time T1 and a threshold T0, and obtaining an early warning probability Pm according to an algebraic relation between the predicted used dose M1 and a threshold M0;
s24: and performing weighted calculation on the early warning probabilities Pt and Pm to obtain the probability P1 of the dose-time sequence early warning model.
4. The ANFIS-based laboratory hazardous chemical anomaly information early warning method according to claim 3, characterized in that: in step S23, the algebraic relationship between the expected return time T1 and the threshold T0 is:
when T1 is less than or equal to T0, the output probability Pt is 0;
when T1 is more than T0 and T1 is less than or equal to 1.2 times of T0, the output probability Pt is 0.2;
when T1 is more than 1.2 times of T0 and T1 is less than or equal to 1.4 times of T0, the output probability Pt is 0.4;
when T1 is more than 1.4 times T0 and T1 is less than or equal to 1.6 times T0, the output probability Pt is 0.6;
when T1 is more than 1.6 times of T0 and T1 is less than or equal to 1.8 times of T0, the output probability Pt is 0.8;
when T1 is more than 1.8 times T0, the output probability Pt is 1.0;
the algebraic relationship between the expected dose M1 and the threshold M0 is:
when M1 is less than or equal to M0, the output probability Pm is 0;
when M1 is more than M0 and M1 is less than or equal to 1.2 times of M0, the output probability Pm is 0.2;
when M1 is more than 1.2 times of M0 and M1 is less than or equal to 1.4 times of M0, the output probability Pm is 0.4;
when M1 is more than 1.4 times of M0 and M1 is less than or equal to 1.6 times of M0, the output probability Pm is 0.6;
when M1 is more than 1.6 times of M0 and M1 is less than or equal to 1.8 times of M0, the output probability Pm is 0.8;
when M1 is more than 1.8 times M0, the output probability Pm is 1.0.
5. The ANFIS-based laboratory hazardous chemical anomaly information early warning method according to claim 3, characterized in that: the weighting calculation formula in step S24 is:
P1＝λ1Pt+(1-λ1)Pm
λ1a weight value in the range of 0-1.
6. The ANFIS-based laboratory hazardous chemical anomaly information early warning method as claimed in claim 1, wherein: the ANFIS early warning model in the step S3 adopts a five-layer fuzzy neural network structure which is respectively a fuzzy layer, a rule layer, a normalization layer, a de-fuzzy layer and an output layer;
all nodes of the fuzzy layer are self-adaptive and are used for fuzzifying all input data according to a membership function, the adopted membership function is a triangular membership function, and the formula is as follows:
where { a, b, c } is a set of preconditions whose values are continuously updated during the training phase, parameters a and c determining the "feet" of the triangle, and parameter b determining the "peaks" of the triangle;
the output function expression of the fuzzification layer is as follows:
OB,j＝μBj(x2)，j＝1,2,3
OA,i、OB,jare respectively data x1，x2At the output of the blurring layer, wherein
And
are respectively data x1，x2A membership function of;
the output of the rule layer multiplies all input data to calculate the excitation intensity w of each rulekThe expression is as follows:
wk＝μA,i(x1)μB,j(x2)，i＝1,2,3,j＝1,2,3
the normalization layer normalizes the excitation intensity of the rule layer, and the calculation formula is as follows:
the de-fuzzy layer is used for calculating a weight value of each node, and the calculation formula is as follows:
wherein p isk、qkAnd kkIs a back-part parameter of the node;
the output layer combines the output of the de-fuzzy layer as the total output, and the calculation formula is as follows:
the output y value is the early warning probability P2 of the ANFIS early warning model.
7. The ANFIS-based laboratory hazardous chemical anomaly information early warning method according to claim 6, wherein: before executing step S3, the method further includes:
acquiring a laboratory hazardous chemicals call record containing predicted return time and predicted usage dosage data for training of an ANFIS network;
training an ANFIS network according to the laboratory dangerous chemical calling record for training to obtain an ANFIS early warning model to be tested;
acquiring a laboratory dangerous chemical calling record containing the predicted return time and the predicted using dosage data for testing the ANFIS network, and acquiring a test result of early warning of abnormal information of the laboratory dangerous chemical by the ANFIS network;
and reading the early warning result of the ANFIS early warning model to the data, comparing training data to obtain the early warning accuracy of the ANFIS early warning model, and taking the ANFIS network as a final ANFIS early warning model.
8. The ANFIS-based laboratory hazardous chemical anomaly information early warning method as claimed in claim 1, wherein: the weighting calculation formula in step S4 is:
P＝λP1+(1-λ)P2
lambda is a weight value with a value range of 0-1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110781857.1A CN113505925B (en) | 2021-07-09 | 2021-07-09 | Laboratory hazardous chemical anomaly information early warning method based on ANFIS |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110781857.1A CN113505925B (en) | 2021-07-09 | 2021-07-09 | Laboratory hazardous chemical anomaly information early warning method based on ANFIS |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113505925A true CN113505925A (en) | 2021-10-15 |
| CN113505925B CN113505925B (en) | 2022-07-15 |
Family
ID=78012702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110781857.1A Active CN113505925B (en) | 2021-07-09 | 2021-07-09 | Laboratory hazardous chemical anomaly information early warning method based on ANFIS |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113505925B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002086814A1 (en) * | 2001-04-23 | 2002-10-31 | Hrl Laboratories, Llc | An on-line method for self-organized learning and extraction of fuzzy rules |
| CN1688244A (en) * | 2002-08-13 | 2005-10-26 | 弗吉尼亚大学专利基金会 | Method, system, and computer program product for the processing of self-monitoring blood glucose(smbg)data to enhance diabetic self-management |
| US20060111857A1 (en) * | 2004-11-23 | 2006-05-25 | Shah Rasiklal P | System and method for predicting component failures in large systems |
| EP2551798A1 (en) * | 2011-07-28 | 2013-01-30 | Fundació CTM Centre Tecnològic | Genetic algorithm-based training of an ANFIS for electric energy consumption forecasting |
| CN103501839A (en) * | 2010-09-09 | 2014-01-08 | S.E.A.医疗系统公司 | Systems and methods for intravenous drug management using immittance spectroscopy |
| CN109242294A (en) * | 2018-08-29 | 2019-01-18 | 国网河南省电力公司电力科学研究院 | Improve the power communication performance method for early warning and device of fuzzy neural network |
| CN109799700A (en) * | 2019-02-20 | 2019-05-24 | 广东工业大学 | A kind of harmful influence transport early warning system based on fuzzy neural network |
| CN109816937A (en) * | 2019-03-29 | 2019-05-28 | 淮阴工学院 | A kind of the gas leakage early warning system and method for tank car |
| CN109878509A (en) * | 2019-03-11 | 2019-06-14 | 东南大学 | Monoblock type tank car Multi-source Information Fusion rollover method for early warning based on fuzzy logic |
| CN110570616A (en) * | 2019-09-10 | 2019-12-13 | 淮阴工学院 | Multipoint fire early warning system based on Internet of things |
| CN111627181A (en) * | 2020-06-28 | 2020-09-04 | 四川旷谷信息工程有限公司 | Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof |
| CN112910859A (en) * | 2021-01-19 | 2021-06-04 | 山西警察学院 | Internet of things equipment monitoring and early warning method based on C5.0 decision tree and time sequence analysis |
-
2021
- 2021-07-09 CN CN202110781857.1A patent/CN113505925B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002086814A1 (en) * | 2001-04-23 | 2002-10-31 | Hrl Laboratories, Llc | An on-line method for self-organized learning and extraction of fuzzy rules |
| CN1688244A (en) * | 2002-08-13 | 2005-10-26 | 弗吉尼亚大学专利基金会 | Method, system, and computer program product for the processing of self-monitoring blood glucose(smbg)data to enhance diabetic self-management |
| US20060111857A1 (en) * | 2004-11-23 | 2006-05-25 | Shah Rasiklal P | System and method for predicting component failures in large systems |
| CN103501839A (en) * | 2010-09-09 | 2014-01-08 | S.E.A.医疗系统公司 | Systems and methods for intravenous drug management using immittance spectroscopy |
| EP2551798A1 (en) * | 2011-07-28 | 2013-01-30 | Fundació CTM Centre Tecnològic | Genetic algorithm-based training of an ANFIS for electric energy consumption forecasting |
| CN109242294A (en) * | 2018-08-29 | 2019-01-18 | 国网河南省电力公司电力科学研究院 | Improve the power communication performance method for early warning and device of fuzzy neural network |
| CN109799700A (en) * | 2019-02-20 | 2019-05-24 | 广东工业大学 | A kind of harmful influence transport early warning system based on fuzzy neural network |
| CN109878509A (en) * | 2019-03-11 | 2019-06-14 | 东南大学 | Monoblock type tank car Multi-source Information Fusion rollover method for early warning based on fuzzy logic |
| CN109816937A (en) * | 2019-03-29 | 2019-05-28 | 淮阴工学院 | A kind of the gas leakage early warning system and method for tank car |
| CN110570616A (en) * | 2019-09-10 | 2019-12-13 | 淮阴工学院 | Multipoint fire early warning system based on Internet of things |
| CN111627181A (en) * | 2020-06-28 | 2020-09-04 | 四川旷谷信息工程有限公司 | Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof |
| CN112910859A (en) * | 2021-01-19 | 2021-06-04 | 山西警察学院 | Internet of things equipment monitoring and early warning method based on C5.0 decision tree and time sequence analysis |
Non-Patent Citations (4)
| Title |
|---|
| KIM, CHAN MOON等: "MLP, ANFIS, and GRNN based real-time coagulant dosage determination and accuracy comparison using full-scale data of a water treatment plant", 《JOURNAL OF WATER SUPPLY:RESEARCH AND TECHNOLOGY》 * |
| 周国荣等: "基于自适应网络模糊推理方法在河道洪水预报中的应用", 《长江流域资源与环境》 * |
| 胡爱国等: "基于ANFIS的地铁杂散电流腐蚀预警系统", 《仪表技术与传感器》 * |
| 黄翰鹏等: "融合模糊神经网络与时序模型的火灾预警算法", 《计算机工程与设计》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113505925B (en) | 2022-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cho | Incorporating soft computing techniques into a probabilistic intrusion detection system | |
| Kaur et al. | Soft computing approaches for prediction of software maintenance effort | |
| Sethukkarasi et al. | An intelligent neuro fuzzy temporal knowledge representation model for mining temporal patterns | |
| CN110088619A (en) | The intelligence system and method for process and assets Gernral Check-up, abnormality detection and control for waste water treatment plant or drinking water plant | |
| US6490573B1 (en) | Neural network for modeling ecological and biological systems | |
| CN110837523A (en) | High-confidence reconstruction quality and false-transient-reduction quantitative evaluation method based on cascade neural network | |
| CN107403188A (en) | A kind of quality evaluation method and device | |
| Ghosh et al. | Bobcat: Bilevel optimization-based computerized adaptive testing | |
| Vlachos | Neuro-fuzzy modeling in bankruptcy prediction | |
| Mahmoud et al. | Network security evaluation using deep neural network | |
| Kumar et al. | Applying soft computing approaches to predict defect density in software product releases: an empirical study | |
| Wang et al. | Statistical inference for networks of high-dimensional point processes | |
| Yovits et al. | Development of a Theory of Information Flow and Analysis | |
| Meskauskas et al. | XAI-based fuzzy SWOT maps for analysis of complex systems | |
| CN113505925B (en) | Laboratory hazardous chemical anomaly information early warning method based on ANFIS | |
| Fatemidokht et al. | Development of a hybrid neuro-fuzzy system as a diagnostic tool for Type 2 Diabetes Mellitus | |
| Reddy et al. | Neural networks for prediction of loan default using attribute relevance analysis | |
| Raqab et al. | On moment generating function of records from extreme value distribution | |
| Lückmann | Simulation-based inference for neuroscience and beyond | |
| CN115565669A (en) | Cancer survival analysis method based on GAN and multitask learning | |
| Alifiah et al. | Prediction of COVID-19 Using the Artificial Neural Network (ANN) with K-Fold Cross-Validation | |
| CN114358186A (en) | Data processing method and device and computer readable storage medium | |
| Brown et al. | Reliability estimation during prototyping of knowledge-based systems | |
| Nord | Multivariate Time Series Data Generation using Generative Adversarial Networks: Generating Realistic Sensor Time Series Data of Vehicles with an Abnormal Behaviour using TimeGAN | |
| Acharya et al. | Quantum Accelerated Causal Tomography: Circuit Considerations Towards Applications |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |