CN112651495A - Neural network circuit for emotion homeostasis adjustment and associative memory - Google Patents

Neural network circuit for emotion homeostasis adjustment and associative memory Download PDF

Info

Publication number
CN112651495A
CN112651495A CN202011483756.8A CN202011483756A CN112651495A CN 112651495 A CN112651495 A CN 112651495A CN 202011483756 A CN202011483756 A CN 202011483756A CN 112651495 A CN112651495 A CN 112651495A
Authority
CN
China
Prior art keywords
module
operational amplifier
voltage
selection switch
gate
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
Application number
CN202011483756.8A
Other languages
Chinese (zh)
Other versions
CN112651495B (en
Inventor
孙军伟
韩俊涛
王春秀
邓秋旺
肖萧
吉浩平
孟子杰
王延峰
王英聪
凌丹
王妍
李盼龙
刘鹏
张勋才
姜素霞
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.)
Zhengzhou University of Light Industry
Original Assignee
Zhengzhou University of Light Industry
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 Zhengzhou University of Light Industry filed Critical Zhengzhou University of Light Industry
Priority to CN202011483756.8A priority Critical patent/CN112651495B/en
Publication of CN112651495A publication Critical patent/CN112651495A/en
Application granted granted Critical
Publication of CN112651495B publication Critical patent/CN112651495B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Neurology (AREA)
  • Circuit For Audible Band Transducer (AREA)

Abstract

The invention provides a neural network circuit for emotion homeostasis adjustment and associative memory, which is used for solving the problem that the conventional neural network circuit does not consider the characteristics of emotion and the influence on memory. The learning device comprises an input signal source, an auditory module, an emotion module and a memory module, wherein the input signal source comprises an excitation signal source and a memory signal source, the excitation signal source is connected with the auditory module, the auditory module and the memory signal source are both connected with the emotion module, the emotion module and the memory signal source are both connected with the memory module, and the memory module outputs a learning completion signal. According to the invention, emotion generation and self-stabilization regulation under different music are realized through the auditory module and the emotion module, learning and memory processes under different emotions are realized, the existing learning and memory mode is expanded, and a more intelligent neural network circuit is favorably constructed; if the auditory voltage is too large, self-protection, namely emotional homeostasis adjustment, is performed, and the emotional voltage protection system is lowered.

Description

Neural network circuit for emotion homeostasis adjustment and associative memory
Technical Field
The invention relates to the technical field of digital-analog circuits and neural network circuits, in particular to a neural network circuit for emotion self-stabilization regulation and associative memory.
Background
Artificial intelligence has since its birth, with the ultimate goal of building intelligent systems that can simulate, and even exceed, the level of human knowledge. The research on the artificial intelligence is helpful for developing intelligent tools and improving the living conditions of people by scientific and technological means. Most current neuromorphic computing devices are based on traditional silicon-based semiconductor materials, but due to moore's law, a new generation of biomimetic devices must be found. The professor California Berkeley, California university, Chucai begonia, in 1971 proposed a resistor-memristor with resistance variability and non-volatility. The Hewlett packard company developed a first titanium oxide-based film memristor physical model in a laboratory in 2008, and confirmed the guess of professor Chua begonia. Memristors not only have nanometer-scale dimensions, but also physical properties very similar to synapses in the smart biological brain. The bioelectronic synapse constructed by the memristor can simulate a simple memory function on the scale of a single neuron, and can be integrated in a large scale to simulate a more complex nervous system.
Memory and emotion are two hot directions in an artificial neural network simulating the functions of a biological brain, but the circuit is difficult to realize due to the complexity of constructing electronic synapses. The advent of memristors has led to the observation that eosins mimic brain function directly at the hardware level. At present, more and more artificial neural network simulation circuit models based on memristors are proposed. Y.v. Pershin et al realized the memory function for the first time with a circuit model of a microcontroller. Zhiging Zeng et al proposed an artificial neural network circuit based on a memristor for babloff associative memory, which realizes a full-function associative memory function. However, the existing neural network circuit based on the memristor only focuses on the memory, and does not consider the characteristics of the emotion and the influence on the memory. The present invention solves this problem.
Disclosure of Invention
Aiming at the technical problems that the characteristics of emotion and the influence on memory are not considered in the conventional neural network circuit, the invention provides the neural network circuit for emotion self-stabilization regulation and associative memory, discusses and simulates emotion self-stabilization which is a special protection mechanism in the brain, and brings emotion into the memory process on the basis of emotion self-stabilization, thereby realizing the learning and memory processes under different emotions.
In order to achieve the purpose, the technical scheme of the invention is realized as follows: a neural network circuit for emotion homeostatic regulation and associative memory comprises an input signal source, an auditory module, an emotion module and a memory module, wherein the input signal source comprises a driving signal source and a memory signal source, the driving signal source is connected with the auditory module, the auditory module and the memory signal source are both connected with the emotion module, the emotion module and the memory signal source are both connected with the memory module, and the memory module outputs a learning completion signal.
The auditory module comprises an auditory signal judgment module and an auditory signal receiving and processing module, an excitation signal source is connected with the auditory signal judgment module, the auditory signal receiving and processing module is connected with the auditory signal receiving and processing module, the auditory signal receiving and processing module is connected with the emotion module, the memory module comprises a memory voltage module and a synapse module, the memory signal source is connected with the memory voltage module, output signals of the memory voltage module and the emotion module are both connected with the synapse module, and an output end of the synapse module outputs a learning completion signal.
The excitation signal source is a music signal, and the music signal comprises an active music signal, a passive music signal and a normal music signal; the memory signal source is a memory signal which comprises a learning signal and a forgetting signal.
The auditory signal receiving and processing module comprises a first auditory signal receiving and processing module, a second auditory signal receiving and processing module and a third auditory signal receiving and processing module, the emotion module comprises a first emotion voltage module, a second emotion voltage module and a third emotion voltage module, the auditory signal judging module is respectively connected with the first auditory signal receiving and processing module, the second auditory signal receiving and processing module is connected with the third auditory signal receiving and processing module, the first auditory signal receiving and processing module and the memory signal are connected with the first emotional voltage module, the second auditory signal receiving and processing module and the memory signal are connected with the second emotional voltage module, the third auditory signal receiving and processing module and the memory signal are connected with the third emotional voltage module, and output signals of the first emotional voltage module, the second emotional voltage module and the third emotional voltage module are connected with the synapse module.
The auditory signal judging module comprises a first operational amplifier, a second operational amplifier and a third operational amplifier, music signals and the first operational amplifier, the non-inverting input ends of the first operational amplifier, the second operational amplifier and the third operational amplifier are respectively connected with three power supplies I, the output end of the first operational amplifier is respectively connected with the first auditory signal receiving and processing module and the first NOT gate, the output ends of the first NOT gate and the second operational amplifier are respectively connected with the first AND gate, the first AND gate is connected with the second auditory signal receiving and processing module, the output end of the second operational amplifier is connected with the second NOT gate, the output ends of the second NOT gate and the third operational amplifier are respectively connected with the second AND gate, and the second AND gate is connected with the third auditory signal receiving and processing module.
The first auditory signal receiving and processing module, the second auditory signal receiving and processing module and the third auditory signal receiving and processing module respectively comprise a first selection switch, a first memristor and a first operational amplifier, the positive pole of the control end of the first selection switch is connected with the output end of the first operational amplifier, the output end of the first AND gate or the output end of the second AND gate, the negative pole of the control end of the first selection switch is grounded, two terminals of the conduction end of the first selection switch are respectively connected with a power supply II and the positive pole of the first memristor, the negative pole of the first memristor is connected with the inverting input end of the first operational amplifier, the inverting input end of the first operational amplifier is connected with the output end of the first operational amplifier through a resistor I, the non-inverting input end of the first operational amplifier is grounded, and the output end of the first operational amplifier outputs an auditory voltage U1Audio voltage U2Or acoustic voltage U3。
The first emotion voltage module and the second emotion voltage module respectively comprise a second selection switch, a first summing device, a second memristor, a third selection switch, a fourth selection switch, a second operational amplifier and a third operational amplifier, and the anode of the control end of the second selection switch is connected with the auditory voltage U1Or acoustic voltage U2The negative pole of the control end of the second selection switch is grounded, two terminals of the conduction end of the second selection switch are respectively connected with the input ends of the power supply II and the first summation device, and the auditory voltage U1Or acoustic voltage U2Connected to the input of a first summing deviceThe output end of the second memristor is connected with the anode of the second memristor, the cathode of the second memristor is connected with the second operational amplifier, the output end of the second operational amplifier is connected with the third operational amplifier, the output ends of the second operational amplifier and the third operational amplifier are both connected with the first logic operation unit of the first emotional voltage module or the second logic module of the second emotional voltage module, the memory signal is connected with the first logic operation unit or the second logic module, the output ends of the first logic operation unit or the second logic power supply are respectively connected with the anodes of the control ends of the third selector switch and the fourth selector switch, the cathodes of the control ends of the third selector switch and the fourth selector switch are grounded, and the output ends of the second operational amplifier and the third operational amplifier are respectively connected with the conduction ends of the third selector switch and the fourth selector switch to obtain the emotional voltage U.4Or emotional voltage U5。
The first logic operation unit comprises a third NOT gate, a third AND gate and a fourth AND gate, the memory signal is connected with the third NOT gate, the output end of the third NOT gate and the output end of the second operational amplifier are both connected with the fourth AND gate, the output end of the second operational amplifier and the memory signal are both connected with the fourth AND gate, the third AND gate is connected with the positive pole of the control end of the third selection switch, and the fourth AND gate is connected with the positive pole of the control end of the fourth selection switch; the second logic operation unit comprises a fourth NOT gate, a fifth AND gate and a sixth AND gate, the memory signal is connected with the fourth NOT gate, the output end of the fourth NOT gate and the output end of the second operational amplifier are both connected with the sixth AND gate, the output end of the second operational amplifier and the memory signal are both connected with the fifth AND gate, the fifth AND gate is connected with the positive pole of the control end of the third selector switch, and the sixth AND gate is connected with the positive pole of the control end of the fourth selector switch.
The third emotion voltage module comprises a fifth selection switch, a second summation device, a fourth operational amplifier and a sixth selection switch, and the auditory voltage U3The negative electrode of the control end of the fifth selection switch is connected with the positive electrode of the control end of the fifth selection switch, the negative electrode of the control end of the fifth selection switch is grounded, and two terminals of the conducting end of the fifth selection switch are respectively connected with a power supply III and a second summing device to listen toSense voltage U3The output end of the second summing device is connected with a fourth operational amplifier, the output end of the fourth operational amplifier is connected with one terminal of the conduction end of a sixth selection switch, the memory signal is connected with the positive pole of the control end of the sixth selection switch, the negative pole of the control end of the sixth selection switch is grounded, and the other terminal of the conduction end of the sixth selection switch outputs an emotional voltage U6。
The memory voltage module comprises a seventh selection switch, an eighth selection switch and a third summation device, a memory signal is connected with the positive pole of the control end of the seventh selection switch, the memory signal is connected with the positive pole of the control end of the eighth selection switch through a fifth NOT gate, the negative poles of the control ends of the seventh selection switch and the eighth selection switch are both grounded, one terminal of the conduction ends of the seventh selection switch and the eighth selection switch is connected with a power supply IV, the other terminal of the conduction ends of the seventh selection switch and the eighth selection switch, and an emotional voltage U4-U6Are connected with a third summing device which is connected with a synapse module; the synapse module comprises a third memristor, a fifth operational amplifier and a voltage processing component ABM1And a sixth operational amplifier, a third summing device is respectively connected with the positive electrode of the third memristor and the voltage processing component ABM1The negative electrode of the third memristor is connected with a fifth operational amplifier, and the fifth operational amplifier is connected with a voltage processing component ABM1Phase connection, voltage processing component ABM1The output end of the first operational amplifier is connected with the inverting input end of the sixth operational amplifier, the non-inverting input end of the sixth operational amplifier is connected with the power supply V, and the output end of the sixth operational amplifier outputs a memory signal.
Compared with the prior art, the invention has the beneficial effects that: emphasis is placed on homeostatic regulation of mood and effects on memory; when the memory signal is input, the circuit will process the input memory signal and the memory module will start to respond. When a music signal is input, the circuit processes the input music signal, the auditory module and the emotion module start to respond, different music signals can generate different emotions, different emotions can generate positive or negative effects on learning, and the influence on learning is more obvious when the emotion intensity is larger; if the auditory voltage generated under the music signal exceeds a threshold, the brain will initiate a protective mechanism, emotional homeostasis, reducing emotional intensity. The invention discusses the self-stabilization regulation of emotion, realizes the learning and memory processes under different emotions on the basis of the self-stabilization of emotion, expands the existing learning and memory mode and is beneficial to constructing a more intelligent neural network circuit. According to the invention, emotion generation and homeostatic regulation under different music are realized through the auditory module and the emotion module, and the memory process under different types of emotions on the basis of emotional homeostatic is discussed through the emotion module and the memory module.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic circuit diagram of the present invention.
Fig. 2 is a circuit diagram of the hearing module shown in fig. 1.
Fig. 3 is a circuit diagram of a first mood voltage module of the mood module of fig. 1.
Fig. 4 is a circuit diagram of a second mood voltage module of the mood module of fig. 1.
Fig. 5 is a circuit diagram of a third mood voltage module of the mood module of fig. 1.
FIG. 6 is a circuit diagram of the memory module shown in FIG. 1.
FIG. 7 is a circuit diagram of the simulation of the memory process under the active music according to the present invention.
FIG. 8 is a circuit diagram of the simulation of the memory process under passive music according to the present invention.
FIG. 9 is a circuit diagram of the memory process simulation under normal music according to the present invention.
In the figure, 1 is a hearing module, 2 is a memory module, 3 is an excitation signal source, 4 is a memory signal source, 5 is an emotion module, 6 is a first selection switch, 7 is a first memristor, 8 is a first operational amplifier, 9 is a second selection switch, 10 is a first summing device, 11 is a second memristor, 12 is a third selection switch, 13 is a fourth selection switch, 14 is a second operational amplifier, 15 is a third operational amplifier, 16 is a fifth selection switch, 17 is a seventh selection switch, 18 is a fourth operational amplifier, 19 is a sixth selection switch, 20 is an eighth selection switch, 21 is a third memristor, 22 is a fifth operational amplifier, and 23 is a sixth operational amplifier.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.
As shown in fig. 1, a neural network circuit for emotion homeostatic regulation and associative memory comprises an input signal source, an auditory module 1, an emotion module 5 and a memory module 2, wherein the input signal source comprises an excitation signal source 3 and a memory signal source 4, the excitation signal source 3 is connected with the auditory module 1, the auditory module 1 and the memory signal source 4 are both connected with the emotion module 5, the emotion module 5 and the memory signal source 4 are both connected with the memory module 2, and the memory module 2 outputs a learning completion signal.
The hearing module 1 comprises a hearing signal judgment module and a hearing signal receiving and processing module, the excitation signal source 3 is connected with the hearing signal judgment module, the hearing signal receiving and processing module is connected with the hearing signal receiving and processing module, the hearing signal receiving and processing module is connected with the emotion module 5, the memory module 2 comprises a memory voltage module and a synapse module, the memory signal source 4 is connected with the memory voltage module, output signals of the memory voltage module and the emotion module 5 are both connected with the synapse module, and an output end of the synapse module outputs a learning completion signal.
The said canThe auditory signal receiving and processing module comprises a first auditory signal receiving and processing module, a second auditory signal receiving and processing module and a third auditory signal receiving and processing module, the emotion module 5 comprises a first emotion voltage module, a second emotion voltage module and a third emotion voltage module, the auditory signal judging module is respectively connected with the first auditory signal receiving and processing module, the second auditory signal receiving and processing module and the third auditory signal receiving and processing module, and the first auditory signal receiving and processing module and the memory signal N are connected2Are all connected with a first emotion voltage module, a second auditory signal receiving and processing module and a memory signal N2Are all connected with a second emotion voltage module, a third auditory signal receiving and processing module and a memory signal N2The first emotional voltage module, the second emotional voltage module and the third emotional voltage module are all connected with a synapse module.
The driving signal source 3 is a music signal N1The music signal comprises an active music signal m1Passive music signal m2And a normal music signal m3(ii) a The memory signal source 4 is a memory signal N2Memorize the signal N2Including a learning signal L1And a forgetting signal F1. Memory signal N2Output learning signal L1The circuit enters a learning state, otherwise the circuit enters a forgetting state. Music signal N1Outputting positive music signal m1And the first auditory signal receiving and processing module and the first emotion voltage module start to respond and generate the positive emotion voltage. Music signal N1Output negative music signal m2The second auditory signal receiving and processing module and the second emotion voltage module start responding to the negative emotion voltage generation. Music signal N1Output normal music signal m3The third auditory signal reception processing module and the third emotional voltage module start responding, but no emotional voltage is generated. The first emotional voltage module, the second emotional voltage module and the third emotional voltage module output voltages to synapse modules in the memory module.
As shown in FIG. 2, the auditory signal determination module comprises a first circuitOperational amplifier OP1A second operational amplifier OP2And a third operational amplifier OP3Music signal N1Are all connected with the first operational amplifier OP1A second operational amplifier OP2And a third operational amplifier OP3Is connected to the non-inverting input terminal of the first operational amplifier OP1A second operational amplifier OP2And a third operational amplifier OP3The three power supplies I are respectively connected with a power supply V1Power supply V2Power supply V3A first operational amplifier OP1And the inverting input terminal of the power supply V1Is connected to the positive pole of a second operational amplifier OP2And the inverting input terminal of the power supply V2Is connected to the positive pole of a third operational amplifier OP3And the inverting input terminal of the power supply V3Is connected with the positive pole of the power supply V1Power supply V2Power supply V3The cathodes of the two are all grounded. Power supply V1Power supply V2Power supply V3The voltage values of (a) are 1.5V, 1.V and 0.5V, respectively. A first operational amplifier OP1A second operational amplifier OP2And a third operational amplifier OP3The function of (a) is to screen for the appropriate voltage and then select the corresponding auditory module. A first operational amplifier OP1A second operational amplifier OP2And a third operational amplifier OP3The voltage of the non-inverting input terminal is greater than that of the inverting input terminal, and then high level is output. A first operational amplifier OP1Respectively connected to the first auditory signal receiving and processing module and the first not gate D1, the first not gate D1 and the second operational amplifier OP2Respectively connected with two input terminals of a first and gate a1, an output terminal of the first and gate a1 connected with a second auditory signal receiving and processing module, and a second operational amplifier OP2Is connected to a second not-gate D2, a second not-gate D2 and a third operational amplifier OP3The output ends of the first and-gate circuits are connected with a second and-gate a2, and the second and-gate a2 is connected with a third auditory signal receiving and processing module.
The first auditory signal receiving and processing module, the second auditory signal receiving and processing module and the third auditory signal receiving and processing moduleThe signal receiving and processing modules respectively comprise a first selection switch 6, a first memristor 7 and a first operational amplifier 8, wherein the positive electrode of the control end of the first selection switch 6 and the first operational amplifier OP1The output end of the first and gate a2 or the output end of the second and gate A3 is connected, the negative electrode of the control end of the first selection switch 6 is grounded, two terminals of the conduction end of the first selection switch 6 are respectively connected with the power supply II and the positive electrode of the first memristor 7, and the first selection switch 6 is used for receiving information of the auditory sense judgment module and further outputting corresponding initial voltage. The negative electrode of the first memristor 7 is connected with the inverting input end of the first operational amplifier 8, the resistance value of the first memristor 7 gradually decreases, the output amplitude of the auditory signal receiving and processing module is larger, and when the resistance value decreases to the minimum value, the output amplitude reaches the peak value of 1V. The inverting input terminal of the first operational amplifier 8 is connected to the output terminal of the first operational amplifier 8 via a resistor I, the non-inverting input terminal of the first operational amplifier 8 is grounded, and the output terminal of the first operational amplifier 8 outputs an acoustic voltage U1Audio voltage U2Or acoustic voltage U3. The first operational amplifier 8 outputs a voltage according to the resistance change of the first memristor 7. In the first auditory signal receiving and processing module, the first selection switch 6 is a selection switch S1Selection switch S1The positive electrode of the control end is connected with a first operational amplifier OP1Power supply II is power supply V4Power supply V4Positive pole of S is connected with a selection switch1The first memristor 7 is a memristor M1The output terminal and the inverting input terminal of the first operational amplifier 8 pass through a resistor R1Connected, the cathode of the first memristor 7 and the operational amplifier OP4Are connected to the inverting input terminal of an operational amplifier OP4The non-inverting input terminal of (2) is grounded, and an operational amplifier OP4Is connected to the inverting input terminal through a resistor R1And operational amplifier OP4Are connected to the output of an operational amplifier OP4Output end of the voltage converter outputs an audible voltage U1. In the second auditory signal receiving and processing module, the first selection switch 6 is a selection switch S5Selection switch S5Is positively connected with the output end of the first AND gate A1Power supply II is power supply V6Power supply V6Positive pole of S is connected with a selection switch5The first memristor 7 is a memristor M3The output terminal and the inverting input terminal of the first operational amplifier 8 pass through a resistor R5Connected, the cathode of the first memristor 7 and the operational amplifier OP7Are connected to the inverting input terminal of an operational amplifier OP7The non-inverting input terminal of (2) is grounded, and an operational amplifier OP7Is connected to the inverting input terminal through a resistor R5And operational amplifier OP7Are connected to the output of an operational amplifier OP7Output end of the voltage converter outputs an audible voltage U2. In the third auditory signal receiving and processing module, the first selection switch 6 is a selection switch S9Selection switch S9The positive pole of the control end is connected with the output end of the first AND gate A2, and the power supply II is a power supply V8Power supply V8Positive pole of S is connected with a selection switch9The first memristor 7 is a memristor M6The output terminal and the inverting input terminal of the first operational amplifier 8 pass through a resistor R9Connected, the cathode of the first memristor 7 and the operational amplifier OP10Are connected to the inverting input terminal of an operational amplifier OP10The non-inverting input terminal of (2) is grounded, and an operational amplifier OP10Is connected to the inverting input terminal through a resistor R9And operational amplifier OP10Are connected to the output of an operational amplifier OP10Output end of the voltage converter outputs an audible voltage U2. Power supply V4Power supply V6Power supply V8The values of (A) are all 1V respectively.
As shown in fig. 3 and 4, each of the first emotion voltage module and the second emotion voltage module includes a second selection switch 9, a first summing device 10, a second memristor 11, a third selection switch 12, a fourth selection switch 13, a second operational amplifier 14, and a third operational amplifier 15, and a control terminal anode of the second selection switch 9 is connected with the auditory voltage U1Or acoustic voltage U2The negative pole of the control end of the second selection switch 9 is grounded, two terminals of the conducting end of the second selection switch 9 are respectively connected with the power supply II and the input end of the first summation device 10, and the second selection switch 9 acts on the input end without the outsideIn case of stimulation, the power source V5And outputting the stable voltage and recovering the resistance value of the second memristor 11. Auditory voltage U1Or acoustic voltage U2The output end of the first summing device 10 is connected with the positive electrode of a second memristor 11, the negative electrode of the second memristor 11 is connected with a second operational amplifier 14, the negative threshold voltage of the second memristor 11 is-0.99V, when the negative threshold voltage is applied to exceed the negative threshold value, the resistance value of the second memristor 11 begins to drop, and then the output voltage of the second operational amplifier 14 is reduced. The output end of the second operational amplifier 14 is connected with a third operational amplifier 15, the output ends of the second operational amplifier 14 and the third operational amplifier 15 are both connected with the first logic operation unit of the first emotion voltage module or the second logic module of the second emotion voltage module, the second operational amplifier 14 outputs corresponding emotion voltage according to the resistance value and the input voltage of the second memristor 11, and the output voltage of the second operational amplifier 14 is negated and output under the action of the third operational amplifier 15. Memory signal N2The output end of the first logic operation unit or the second logic power supply is respectively connected with the control ends of the third selection switch 12 and the fourth selection switch 13, the control ends of the third selection switch 12 and the fourth selection switch 13 are grounded, the output ends of the second operational amplifier 14 and the third operational amplifier 15 are respectively connected with the conducting ends of the third selection switch 12 and the fourth selection switch 13 to obtain the emotional voltage U4Or emotional voltage U5. The third selector switch 12 and the fourth selector switch 13 function.
As shown in FIG. 3, the second selection switch 9 of the first mood voltage module is a selection switch S2Power supply II is power supply V5Selection switch S2One terminal of the conducting end and the power supply V5Is connected with the positive pole of the power supply V5Negative pole of (3) is grounded, and the hearing voltage U1When less than zero, the selection switch S2Is turned on, otherwise the switch S is selected2And (5) disconnecting. The first summing device 10 is a summing device SUM1, and the second memristor 11 is a memristor M2Third selection switch12 is a selector switch S3The fourth selector switch 13 is a selector switch S4. The second operational amplifier 14 comprises an operational amplifier OP5And a resistance R2An operational amplifier OP5The non-inverting input terminal of (2) is grounded, and an operational amplifier OP5Respectively connected with resistors R2And memristor M2Negative electrode, resistance R2Connecting operational amplifiers OP5Of an operational amplifier OP5Respectively connected with a selection switch S3One terminal and a third operational amplifier 15. The third operational amplifier 15 comprises an operational amplifier OP6Resistance R3And a resistance R4An operational amplifier OP5Is connected with the resistor R3Connected, operational amplifier OP6The non-inverting input end of the resistor R is grounded3Respectively connected with resistors R4And operational amplifier OP5The inverting input terminal of (3), the resistor R4Connecting operational amplifiers OP6Of an operational amplifier OP6Is connected with the selection switch S4One terminal of the conducting terminal. Selection switch S3And a selector switch S4The other terminal of the conducting end outputs an output voltage U4. As shown in FIG. 4, the second selection switch 9 of the second mood voltage module is a selection switch S6Power supply II is power supply V7Selection switch S6One terminal of the conducting end and the power supply V7Is connected with the positive pole of the power supply V7The first summing device 10 is a summing device SUM2, and the second memristor 11 is a memristor M4The third selection switch 12 is a selection switch S7The fourth selector switch 13 is a selector switch S8Acoustic voltage U2When less than zero, the selection switch S6Is turned on, otherwise the switch S is selected6And (5) disconnecting. The second operational amplifier 14 comprises an operational amplifier OP8And a resistance R6An operational amplifier OP8The non-inverting input terminal of (2) is grounded, and an operational amplifier OP8Respectively connected with resistors R6And memristor M4Negative electrode, resistance R6Connecting operational amplifiers OP8Output terminal of (1), operational amplificationDevice OP8Respectively connected with a selection switch S7One terminal and a third operational amplifier 15. The third operational amplifier 15 comprises an operational amplifier OP9Resistance R7And a resistance R8An operational amplifier OP8Is connected with the resistor R7Connected, operational amplifier OP9The non-inverting input end of the resistor R is grounded7Respectively connected with resistors R8And operational amplifier OP9The inverting input terminal of (3), the resistor R8Connecting operational amplifiers OP9Of an operational amplifier OP9Is connected with the selection switch S8One terminal of the conducting terminal. Selection switch S7And a selector switch S8The other terminal of the conducting end outputs an output voltage U5. Power supply V5And a power supply V7The value of (1) is 0.5V, and the voltage is recovered.
As shown in FIG. 3, the first logic unit is used for controlling the selection switch S according to the input signal3And a selector switch S4To open and close. When the operational amplifier OP5When outputting positive voltage, if N is2Output high level, NOT gate D3Output low level, AND gate A4Output high level, select switch S4Conducting; when the operational amplifier OP5When outputting positive voltage, if N is2Output low level, NOT gate D3Output high level, AND gate A3Output high level, select switch S3And conducting. The first logic operation unit comprises a third NOT gate D3And a third AND gate A3And a fourth AND gate A4Memorize the signal N2And a third not gate D3Connected, third not gate D3And the output of the second operational amplifier 14 and the fourth and gate a4Connected to the output of the second operational amplifier 14 and the memory signal N2Are all connected with a fourth AND gate A4Connected, the third AND gate A3A fourth AND gate A connected with the control end anode of the third selection switch 124Is connected with the positive pole of the control end of the fourth selection switch 13; auditory voltage U1When not zero, the third AND gate A3Fourth AND gate A4The first input terminal of the first memory cell has a high input voltage and memorizes a signal N2When outputting high level, the selection switch S4Conducting, memorizing signal N2When outputting low level, the selection switch S3And conducting.
As shown in FIG. 4, the second logic operation unit is used for controlling the selection switch S according to the input signal7And a selector switch S8To open and close. When the operational amplifier OP8When outputting positive voltage, if N is2Output high level, NOT gate D4Output low level, AND gate A5Output high level, select switch S7Conducting; when the operational amplifier OP8When outputting positive voltage, if N is2Output low level, NOT gate D4Output high level, AND gate A6Output high level, select switch S8And conducting. The second logic operation unit comprises a fourth NOT gate D4Fifth and gate A5And a sixth AND gate A6Memorize the signal N2And a fourth not gate D4Connected, fourth not gate D4And the output of the second operational amplifier 14 and the sixth and gate a6Connected to the output of the second operational amplifier 14 and the memory signal N2Are all connected with a fifth AND gate A5Connected, the fifth AND gate A5A sixth AND gate A connected with the control end anode of the third selection switch 126And is connected to the positive terminal of the control terminal of the fourth selector switch 13. Auditory voltage U2When not zero, AND gate A5、A6The first input terminal of the first memory cell has a high input voltage and memorizes a signal N2When outputting high level, the selection switch S7Conducting, memorizing signal N2When outputting low level, the selection switch S8And conducting.
As shown in fig. 5, the third emotional voltage module includes a fifth selection switch 16, a second summing device SUM3A fourth operational amplifier 18 and a sixth selection switch 19, the fifth selection switch 16 being a selection switch S10Acoustic voltage U3A control terminal anode of the fifth selection switch 16 is connected to a control terminal cathode of the fifth selection switch 16 and grounded, and a conduction terminal of the fifth selection switch 16Two terminals of which are respectively connected with the positive pole of the power supply III, i.e. the power supply V9, and the second summing device SUM3Connected, the negative pole of the power supply V9 is grounded, and the auditory voltage U is3When less than zero, the selection switch S10Is turned on, otherwise the switch S is selected10And (5) disconnecting. Auditory voltage U3And a second summing device SUM3Phase-connected, second summing device SUM3Is connected to a fourth operational amplifier 18, the fourth operational amplifier 18 being responsive to a resistor R10And the corresponding emotional voltage is output by the resistance value of the resistor and the input voltage. The output of the fourth operational amplifier 18 and a sixth selection switch 19, i.e. selection switch S11One terminal of the conducting terminal is connected to memorize the signal N2The positive pole of the control end of the sixth selection switch 19 is connected, the negative pole of the control end of the sixth selection switch 19 is grounded, and the other terminal of the conduction end of the sixth selection switch 19 outputs the emotional voltage U6. Auditory voltage U3Non-zero, signal N is memorized2When outputting high level, the selection switch S11Conducting, memorizing signal N2When outputting low level, the selection switch S11And (5) disconnecting. The fourth operational amplifier 18 comprises an operational amplifier OP11Resistance R10And a resistance R11Resistance R10And a second summing device SUM3Connected with the other end respectively with an operational amplifier OP11And the resistor R11Resistance R11And operational amplifier OP11Are connected to the output of an operational amplifier OP11Is connected with the selection switch S11One terminal of the conducting terminal.
As shown in FIG. 6, the memory voltage module includes a seventh selection switch 17, i.e., a selection switch S12The eighth selection switch 20, i.e., the selection switch S13And a third summing device SUM4Memorize the signal N2A control terminal of the seventh selection switch 17 is connected with the positive electrode, and the signal N is memorized2Through a fifth NOT gate D5The control end positive pole of the eighth selection switch 20 is connected, the control end negative poles of the seventh selection switch 17 and the eighth selection switch 20 are both grounded, and one terminal of the conducting end of the seventh selection switch 17 and the eighth selection switch 20 is connected with the power supply IVConnected to one terminal of the conducting terminal of the seventh selection switch 17 and to the power supply V10Is connected with the positive pole of the power supply V10The negative electrode of (2) is grounded. One terminal of the conduction terminal of the eighth selection switch 20 and the power supply V11Is connected with the positive pole of the power supply V11The negative electrode of (2) is grounded. The other terminals of the conducting terminals of the seventh selection switch 17 and the eighth selection switch 20, and the emotional voltage U4-U6SUM and third summing device SUM4Connected, the output voltages of the three mood modules are not output at the same time, and a third summing device SUM4And 2 voltage signals are superposed. Third summing device SUM4Connecting with a synapse module; memory signal N2When outputting high level, the selection switch S12Conducting, memorizing signal N2When outputting low level, the selection switch S13And conducting. The synaptic module comprises a third memristor 21, i.e. memristor M6A fifth operational amplifier 22, a voltage processing component ABM1And a fifth operational amplifier 23, a third summing device SUM4Respectively connected with the anode of the third memristor 21 and the voltage processing component ABM1The negative pole of a third memristor 21 is connected with a fifth operational amplifier 22, and a memristor M6The resistance value of (1) is reduced under positive voltage and increased under negative voltage, and the memristor M6After the resistance value of (1) falls to the threshold value of 1.8K, the fifth operational amplifier 23 outputs a high level. Fifth operational amplifier 22 and voltage processing component ABM1Phase connection, voltage processing component ABM1Is connected with the inverting input terminal of the sixth operational amplifier 23, the non-inverting input terminal of the sixth operational amplifier 23 is connected with the power supply V, and the output terminal of the sixth operational amplifier 23 outputs the memory signal N3I.e. a learning complete signal. Voltage processing component ABM1Has an output value of-IN2/IN1=M1/1000, the voltage summing device of the invention model SUM1-SUM4 is SUM50N03, operational amplifier OP1-OP13The model is LM 393.
The fifth operational amplifier 22 comprises an operational amplifier OP12An operational amplifier OP12The non-inverting input terminal of (2) is grounded, and an operational amplifier OP12Respectively connected with the resistorR12And memristor M6Is connected to the negative pole of the resistor R12And operational amplifier OP12Are connected to the output of an operational amplifier OP12Output end and voltage processing component ABM1Is connected to one input terminal of the operational amplifier OP of the sixth operational amplifier 2313The power supply V is a power supply V12An operational amplifier OP13Power supply V connected with non-inverting input terminal12Positive electrode, power supply V12The negative electrode is grounded.
The memory signal N2Output high level, the circuit enters learning state, select switch S12Conducting, memorizing the voltage module to output voltage V to synapse module10(ii) a Memory signal N2Output low level, circuit enter forget state, select switch S13Conducting, memorizing the voltage module to output voltage V to synapse module11. Music signal N1The output voltage of 1.5V, 1V and 0.5V respectively represent positive music signal m1Passive music signal m2And a normal music signal m3Memorize the signal N2Respectively correspond to the learning signals L1And a forgetting signal F1. The auditory signal judgment module is responsible for identifying the music signal N1A first auditory signal reception processing module, a second auditory signal reception processing module and a third auditory signal reception processing module respectively process the positive music signal m1Passive music signal m2And a normal music signal m3The output voltages of the first auditory signal receiving and processing module, the second auditory signal receiving and processing module and the third auditory signal receiving and processing module are respectively U1、U2And U3The first emotion voltage module, the second emotion voltage module and the third emotion voltage module of the emotion module respectively process the voltage U1、U2And U3And respectively outputs voltage U to synapse module in memory module4、U5And U6. Output voltages of the first emotional voltage module, the second emotional voltage module and the third emotional voltage module can influence normal memristor M6The resistance value of (a) is increased and decreased,i.e. affect the normal memory process. According to the actual situation, the positive music signal m1Positive emotion can be generated, so that the learning speed is promoted, and the forgetting speed is reduced; negative music signal m2Negative emotion can be generated, so that the learning speed is reduced, and the forgetting speed is promoted; normal music signal m3The emotion can not be generated, and the learning and forgetting speed is hardly influenced.
Specifically, the circuit structure of the invention can realize the learning and memory functions of people under different music on the basis of emotional homeostasis. Memory signal N2And outputting a high level and entering a learning mode. Learning for T seconds under normal condition, the synaptic weight of the synaptic module is reduced to a threshold value, and a signal N is memorized3And outputting a high level, namely finishing learning. Memory process under active music, memory signal N2Outputting a high level and entering a learning mode; music signal N1Outputting positive music signal m1The first auditory signal receiving and processing module generates an auditory voltage U1The first emotion voltage module receives the auditory voltage U1And generates an emotional voltage U4Emotional voltage U4For memristor M in synaptic module6The applied voltage of the first and second modules has an influence, learning is carried out for W seconds under the condition of positive music, the synaptic weight of the synaptic module is reduced to a threshold value, and a signal N is memorized3And outputting a high level, namely finishing learning. Memory process under passive music, memory signal N2Outputting a high level and entering a learning mode; music signal N1Output negative music signal m2The second auditory signal receiving and processing module generates an auditory voltage U2The second emotion voltage module receives the auditory voltage U2And generates an emotional voltage U5Emotional voltage U4For memristor M in synaptic module6The applied voltage of the first and second modules has an influence, learning is carried out for X seconds under negative music, the synaptic weight of the synaptic module is reduced to a threshold value, and a signal N is memorized3And outputting a high level, namely finishing learning. Memory process under normal music, memory signal N2Outputting a high level and entering a learning mode; music signal N1Output normal music signal m2Third auditory signal receiving and processing moduleGenerating an auditory voltage U3The third emotion voltage module receives the auditory voltage U3And generates an emotional voltage U6When learning is performed for Z seconds under normal music, the synaptic weight of the synaptic module is reduced to a threshold value, and a signal N is memorized3And outputting a high level, namely finishing learning.
In particular, the first step of the memorization process under positive music, the signal N is memorized2Output high level, select switch S12On, power supply V10Voltage-summed device SUM4To the memristor M6Positive electrode of (1), memristor M6The resistance value begins to decrease and enters a learning state; music signal N1Outputting positive music signal m1An operational amplifier OP1Output high level, first AND gate A1And a second AND gate A2Output low level, select switch S1On, power supply V4Output voltage, memristor M1The resistance value begins to decrease, and the memristor M1Is applied with a voltage via an operational amplifier OP4Post-action output auditory voltage U1. Auditory voltage U1As memristors M2Input voltage of, memristor M1Is applied with a voltage via an operational amplifier OP4And outputting voltage after acting. Operational amplifier OP5Output voltage, third AND gate A3Output low level, fourth A4Output high level, select switch S4Conducting, operational amplifier OP5Is output voltage via an operational amplifier OP6SUM (summation device) for negation and backward summation4Second input terminal of the voltage regulator outputs an emotional voltage U4. When the operational amplifier OP4Output voltage of (2) exceeding memristor M1At threshold value of, i.e. acoustic voltage U1Oversized memristor M1The resistance value of (1) is increased, namely, the emotional self-stabilization effect and the emotional voltage U occur4The value of (c) decreases. Memristor M in synapse module6Applying a voltage of the summing device SUM4Of the output voltage, i.e. the power supply V10Voltage of plus emotional voltage U4(ii) a Memristor M6The resistance value begins to drop, and the voltage processing component ABM1Is equal in value toMemristor M6Resistance value of, as a voltage processing component ABM1Is less than the power supply V12At a voltage of (4), the operational amplifier OP13And outputting high level and finishing learning. The second step of the memory process under the active music, the memory signal N2Output low level, select switch S13On, power supply V11Voltage summed device SUM4To the memristor M6Positive electrode of (1), memristor M6The resistance value begins to rise, and the system enters a forgetting state. Music signal N1Outputting positive music signal m1An operational amplifier OP1Output high level, first AND gate A1And a second AND gate A2Output low level, select switch S1On, power supply V4Output voltage, memristor M1The resistance value begins to decrease, and the memristor M1Is applied with a voltage via an operational amplifier OP4Post-action output auditory voltage U1. Auditory voltage U1As memristors M2Input voltage of, memristor M1Is applied with a voltage via an operational amplifier OP4After applied, output voltage, third AND gate A3Output high level, fourth AND gate A4Output low level, select switch S3Conducting, operational amplifier OP4Output voltage radial summation device SUM4Second input terminal of the voltage regulator outputs an emotional voltage U4. Memristor M in synapse module6Applying a voltage of the summing device SUM4Of the output voltage, i.e. the power supply V11Voltage of plus emotional voltage U4Memristor M6The resistance value of (A) begins to rise, and the voltage processing component ABM1The output voltage of (1) is equal to the memristor M in value6When ABM is reached1Is greater than the power supply V12At a voltage of (4), the operational amplifier OP13The output is low level and the memory is forgotten. Positive music signal m1Disappearing and memory resistor M1Automatically restores to the initial value, selects the switch S2On, power supply V5Output voltage, memristor M2The resistance value of (a) starts to decrease and returns to the initial value. Under positive music signal, generatingThe emotion is positive, the learning speed is accelerated, and the forgetting speed is reduced. Fig. 7 is a simulation waveform of the memory process under the positive music.
In particular, the first step of the memorization process under passive music, the signal N is memorized2Output high level, select switch S12On, power supply V10Voltage-summed device SUM4To the memristor M6Positive electrode of (1), memristor M6The resistance value begins to decrease and enters a learning state; music signal N1Outputting positive music signal m2An operational amplifier OP1Output low level, first AND gate A1Output high level, second AND gate A2Output low level, select switch S5On, power supply V6Output voltage, memristor M3The resistance value begins to decrease, and the memristor M3Is applied with a voltage via an operational amplifier OP7Post-action output auditory voltage U2. Auditory voltage U2As memristors M3Input voltage of, memristor M3Is applied with a voltage via an operational amplifier OP7A sixth AND gate A for outputting voltage after being applied6Output low level, fifth AND gate A5Output high level, select switch S7Conducting, operational amplifier OP8SUM-of-direction device SUM4Second input terminal of the voltage regulator outputs an emotional voltage U5. When the operational amplifier OP7Output voltage of (2) exceeding memristor M4At threshold value of, i.e. acoustic voltage U1Oversized memristor M4Rise in resistance, i.e. steady-state effect, emotional voltage U5The value of (c) decreases. Memristor M in synapse module6Applying a voltage of the summing device SUM4Of the output voltage, i.e. the power supply V10Voltage of plus emotional voltage U5Memristor M6The resistance value begins to drop, and the voltage processing component ABM1The output voltage of (1) is equal to the memristor M in value6When ABM is reached1Is less than the power supply V12At a voltage of (4), the operational amplifier OP13And outputting high level and finishing learning. A second step of the memorization process under passive music, the memorization signal N2Output low electricityFlat, select switch S13On, power supply V11Voltage summed device SUM4To the memristor M6Positive electrode of (1), memristor M6The resistance value of the capacitor starts to rise and enters a forgetting state. Music signal N1Outputting positive music signal m2Selection switch S1On, power supply V4Output voltage, memristor M3The resistance value begins to decrease, and the memristor M3Is applied with a voltage via an operational amplifier OP7Post-action output auditory voltage U2. Auditory voltage U2As memristors M4Input voltage of, memristor M3Is applied with a voltage via an operational amplifier OP7A sixth AND gate A for outputting voltage after being applied6Output high level, fifth AND gate A5Output low level, select switch S8Conducting, operational amplifier OP8Is output voltage via an operational amplifier OP8SUM (post-action output Voltage summing device)4Second input terminal of the voltage regulator outputs an emotional voltage U5. Memristor M in synapse module6Applying a voltage of the summing device SUM4Of the output voltage, i.e. the power supply V11Voltage of plus emotional voltage U5Memristor M6The resistance value of (A) begins to rise, and the voltage processing component ABM1The output voltage of (1) is equal to the memristor M in value6When ABM is reached1Is greater than the power supply V12At a voltage of (4), the operational amplifier OP13The output is low level and the memory is forgotten. Negative music signal m2Disappearing and memory resistor M3Automatically restores to the initial value, selects the switch S6On, power supply V7Output voltage, memristor M4The resistance value of (a) starts to decrease and returns to the initial value. Under the negative music signal, the passive emotion is generated, the learning speed is reduced, and the forgetting speed is increased. Fig. 8 is a waveform of memory process simulation under passive music.
Specifically, the first step of the memory process under normal music is to memorize the signal N2Output high level, select switch S12On, power supply V10Voltage-summed device SUM4First input ofTerminal output to memristor M6Positive electrode of (1), memristor M6The resistance value of (1) starts to decrease, and the learning state is entered. Music signal N1Outputting positive music signal m3An operational amplifier OP1Output low level, first AND gate A1Output low level, second AND gate A2Output high level, select switch S9On, power supply V8Output voltage, memristor M5The resistance value begins to decrease, and the memristor M5Is applied with a voltage via an operational amplifier OP10Post-action output auditory voltage U3. Auditory voltage U3As an input voltage of the third emotional voltage module, unlike other emotional voltage modules, a memristor in the third emotional voltage module is replaced with a large resistance R10I.e. whether R is10How the applied voltage is changed, the operational amplifier OP10Is close to zero, i.e. the output voltage U of the third emotional voltage module6Almost has no influence on the whole memory process, and does not generate emotional homeostasis. The second step of the memory process under normal music, memory signal N2Output low level, select switch S13On, power supply V11Voltage summed device SUM4To the memristor M6Positive electrode of (1), memristor M6The resistance value of the capacitor starts to rise and enters a forgetting state. Fig. 9 is a waveform of memory process simulation under normal music.
According to the neural network circuit for emotion self-stabilization regulation and associative memory, when different music signals are input into the circuit, corresponding auditory voltage can be generated, emotion voltage is further generated, if the auditory voltage is too large, self-protection can be performed on the system, namely emotion self-stabilization regulation is performed, the emotion voltage is reduced, and the system is protected. Meanwhile, the influence of different types of emotions on the memory process is discussed on the basis of emotional homeostasis, and the existing memory mode is expanded.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (10)

1. The utility model provides a neural network circuit towards regulation of mood homeostasis and associative memory, characterized in that, including input signal source, auditory module (1), mood module (5) and memory module (2), input signal source includes excitation signal source (3) and memory signal source (4), excitation signal source (3) are connected with auditory module (1), auditory module (1) and memory signal source (4) all are connected with mood module (5), mood module (5) and memory signal source (4) all are connected with memory module (2), memory module (2) output learning completion signal.
2. The neural network circuit oriented to emotion homeostatic regulation and associative memory according to claim 1, wherein the auditory module (1) includes an auditory signal judging module and an auditory signal receiving and processing module, the excitation signal source (3) is connected to the auditory signal judging module, the auditory signal receiving and processing module is connected to the auditory signal receiving and processing module, the auditory signal receiving and processing module is connected to the emotion module (5), the memory module (2) includes a memory voltage module and a synapse module, the memory signal source (4) is connected to the memory voltage module, output signals of the memory voltage module and the emotion module (5) are both connected to the synapse module, and an output end of the synapse module outputs a learning completion signal.
3. The neural network circuit for emotional homeostatic regulation and associative memory according to claim 2, characterized in that said excitation signal source (3) is a music signal (N)1) The music signals include positive music signals, negative music signals and normal music signals; the memory signal source (4) is a memory signal (N)2) Memorize the signal (N)2) Including a learning signal and a forgetting signal.
4. The neural network circuit for emotional homeostasis modulation and associative memory according to claim 2 or 3, whereinCharacterized in that the auditory signal receiving and processing module comprises a first auditory signal receiving and processing module, a second auditory signal receiving and processing module and a third auditory signal receiving and processing module, the emotion module (5) comprises a first emotion voltage module, a second emotion voltage module and a third emotion voltage module, the auditory signal judging module is respectively connected with the first auditory signal receiving and processing module, the second auditory signal receiving and processing module and the third auditory signal receiving and processing module, and the first auditory signal receiving and processing module and the memory signal (N) are connected2) Are all connected with a first emotion voltage module, a second auditory signal receiving and processing module and a memory signal (N)2) Are all connected with a second emotion voltage module, a third auditory signal receiving and processing module and a memory signal (N)2) The first emotional voltage module, the second emotional voltage module and the third emotional voltage module are all connected with a synapse module.
5. The neural network circuit for emotional homeostasis modulation and associative memory according to claim 4, wherein the auditory signal determination module comprises a first operational amplifier (OP)1) A second operational amplifier (OP)2) And a third operational amplifier (OP)3) Music signal (N)1) Are all connected with a first operational amplifier (OP)1) A second operational amplifier (OP)2) And a third operational amplifier (OP)3) Is connected to the non-inverting input of a first operational amplifier (OP)1) A second operational amplifier (OP)2) And a third operational amplifier (OP)3) Are respectively connected with three power supplies I, a first operational amplifier (OP)1) Respectively connected with the first auditory signal receiving and processing module and the first NOT gate (D1), the first NOT gate (D1) and the second operational amplifier (OP)2) The output ends of the first and-gate circuits are connected with a first and-gate (A1), the first and-gate (A1) is connected with a second auditory signal receiving and processing module, and a second operational amplifier (OP)2) Is connected to a second not-gate (D2), aA two not gate (D2) and a third operational amplifier (OP)3) The output ends of the first and-gate modules are connected with a second and-gate (A2), and the second and-gate (A2) is connected with a third auditory signal receiving and processing module.
6. The neural network circuit oriented to emotion homeostatic regulation and associative memory according to claim 5, wherein the first, second and third auditory signal receiving and processing modules each comprise a first selection switch (6), a first memristor (7) and a first operational amplifier (8), and the positive electrode of the control terminal of the first selection switch (6) and the first operational amplifier (OP) are connected in series1) The output end of the first operational amplifier (8), the output end of the first AND gate (A2) or the output end of the second AND gate (A3) is connected, the negative electrode of the control end of the first selection switch (6) is grounded, two terminals of the conduction end of the first selection switch (6) are respectively connected with a power supply II and the positive electrode of a first memristor (7), the negative electrode of the first memristor (7) is connected with the inverting input end of the first operational amplifier (8), the inverting input end of the first operational amplifier (8) is connected with the output end of the first operational amplifier (8) through a resistor I, the non-inverting input end of the first operational amplifier (8) is grounded, and the output end of the first operational amplifier (8) outputs an auditory voltage U1Audio voltage U2Or acoustic voltage U3。
7. The neural network circuit oriented to emotional homeostatic regulation and associative memory according to claim 1 or 6, wherein the first emotional voltage module and the second emotional voltage module each comprise a second selection switch (9), a first summing device (10), a second memristor (11), a third selection switch (12), a fourth selection switch (13), a second operational amplifier (14) and a third operational amplifier (15), and the positive electrode of the control terminal of the second selection switch (9) is connected with the auditory voltage Up1Or acoustic voltage U2The negative pole of the control end of the second selection switch (9) is grounded, and two terminals of the conduction end of the second selection switch (9) are respectively connected with the power supply II and the first summing device(10) Is connected to the input terminal of the hearing voltage U1Or acoustic voltage U2The memory signal is connected with the input end of a first summing device (10), the output end of the first summing device (10) is connected with the positive electrode of a second memristor (11), the negative electrode of the second memristor (11) is connected with a second operational amplifier (14), the output end of the second operational amplifier (14) is connected with a third operational amplifier (15), the output ends of the second operational amplifier (14) and the third operational amplifier (15) are connected with a first logic operation unit of a first emotion voltage module or a second logic module of a second emotion voltage module, and a memory signal (N)2) The output end of the first logic operation unit or the second logic power supply is respectively connected with the control ends of the third selection switch (12) and the fourth selection switch (13) in an anode way, the control ends of the third selection switch (12) and the fourth selection switch (13) are grounded in a cathode way, the output ends of the second operational amplifier (14) and the third operational amplifier (15) are respectively connected with the conduction ends of the third selection switch (12) and the fourth selection switch (13) in a conduction way, and the emotional voltage U is obtained4Or emotional voltage U5。
8. The neural network circuit for emotional homeostasis modulation and associative memory according to claim 7, wherein the first logic operation unit comprises a third not gate (D)3) And a third AND gate (A)3) And a fourth AND gate (A)4) Memorize the signal (N)2) And a third not gate (D)3) Connected, the third not gate (D)3) And the output of the second operational amplifier (14) and a fourth AND gate (A)4) Connected, the output of the second operational amplifier (14) and the memory signal (N)2) Are all connected with a fourth AND gate (A)4) Connected to a third AND gate (A)3) A fourth AND gate (A) connected with the positive pole of the control end of the third selection switch (12)4) The positive pole of the control end of the fourth selection switch (13) is connected; the second logic operation unit includes a fourth not gate (D)4) Fifth and fifthDoor (A)5) And a sixth AND gate (A)6) Memorize the signal (N)2) And a fourth not gate (D)4) Connected, fourth not gate (D)4) And the output of the second operational amplifier (14) and a sixth AND gate (A)6) Connected, the output of the second operational amplifier (14) and the memory signal (N)2) Are all connected with a fifth AND gate (A)5) Connected with a fifth AND gate (A)5) A sixth AND gate (A) connected with the control end anode of the third selection switch (12)6) And is connected with the positive pole of the control end of the fourth selection switch (13).
9. The neural network circuit oriented to emotional homeostasis adjustment and associative memory according to claim 4, wherein the third emotional voltage module comprises a fifth selection switch (16), a second summing device (SUM)3) A fourth operational amplifier (18) and a sixth selection switch (19), an auditory voltage U3The negative pole of the control end of the fifth selection switch (16) is connected with the positive pole of the control end of the fifth selection switch (16) and grounded, and two terminals of the conducting end of the fifth selection switch (16) are respectively connected with a power supply III and a second summation device (SUM)3) Connected, hearing voltage U3And a second summing device (SUM)3) Phase-connected, second summing device (SUM)3) Is connected to a fourth operational amplifier (18), the output of the fourth operational amplifier (18) is connected to one terminal of the conduction terminal of a sixth selection switch (19), a memory signal (N) is stored2) The positive pole of the control end of the sixth selection switch (19) is connected, the negative pole of the control end of the sixth selection switch (19) is grounded, and the other terminal of the conduction end of the sixth selection switch (19) outputs an emotional voltage U6。
10. The neural network circuit oriented to emotional homeostatic regulation and associative memory according to claim 8 or 9, wherein the memory voltage module comprises a seventh selection switch (17), an eighth selection switch (20) and a third summation device (SUM)4），Memory signal (N)2) A control terminal positive electrode connected to the seventh selection switch (17) for memorizing the signal (N)2) Through a fifth NOT gate (D)5) The positive pole of the control end of the eighth selection switch (20) is connected, the negative poles of the control ends of the seventh selection switch (17) and the eighth selection switch (20) are grounded, one terminal of the conduction ends of the seventh selection switch (17) and the eighth selection switch (20) is connected with a power supply IV, and the other terminal of the conduction ends of the seventh selection switch (17) and the eighth selection switch (20) and an emotional voltage U are connected4-U6Are all combined with a third summing device (SUM)4) Phase-connected, third summing device (SUM)4) Connecting with a synapse module; the synapse module comprises a third memristor (21), a fifth operational amplifier (22) and a voltage processing component ABM1And a sixth operational amplifier (23), a third summing device (SUM)4) Respectively connected with the anode of the third memristor (21) and a voltage processing component ABM1The negative electrode of the third memristor (21) is connected with a fifth operational amplifier (22), and the fifth operational amplifier (22) is connected with a voltage processing component ABM1Phase connection, voltage processing component ABM1Is connected with the inverting input terminal of a sixth operational amplifier (23), the non-inverting input terminal of the sixth operational amplifier (23) is connected with a power supply V, and the output terminal of the sixth operational amplifier (23) outputs a memory signal (N)3）。
CN202011483756.8A 2020-12-16 2020-12-16 Neural network circuit for emotion homeostasis adjustment and associative memory Active CN112651495B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011483756.8A CN112651495B (en) 2020-12-16 2020-12-16 Neural network circuit for emotion homeostasis adjustment and associative memory

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011483756.8A CN112651495B (en) 2020-12-16 2020-12-16 Neural network circuit for emotion homeostasis adjustment and associative memory

Publications (2)

Publication Number Publication Date
CN112651495A true CN112651495A (en) 2021-04-13
CN112651495B CN112651495B (en) 2021-10-12

Family

ID=75354197

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011483756.8A Active CN112651495B (en) 2020-12-16 2020-12-16 Neural network circuit for emotion homeostasis adjustment and associative memory

Country Status (1)

Country Link
CN (1) CN112651495B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113344194A (en) * 2021-05-31 2021-09-03 郑州轻工业大学 Operational conditioned reflex circuit based on memristor
CN113658493A (en) * 2021-08-20 2021-11-16 安徽大学 Reinforced learning bionic circuit architecture for simulating associative memory
CN114997387A (en) * 2022-06-28 2022-09-02 郑州轻工业大学 Visual-evoked emotional habituation characteristic neural network circuit
CN117610629A (en) * 2023-12-20 2024-02-27 湖南师范大学 Memristive neural network circuit for inhibiting negative emotion and self-repairing based on bionic mechanism

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106815636A (en) * 2016-12-30 2017-06-09 华中科技大学 A kind of neuron circuit based on memristor
CN110110840A (en) * 2019-04-22 2019-08-09 中国地质大学（武汉） A kind of associative memory emotion recognition circuit based on memristor neural network
US20200005130A1 (en) * 2018-07-02 2020-01-02 Kabushiki Kaisha Toshiba Reinforcement learning system
CN110910723A (en) * 2019-12-23 2020-03-24 郑州轻工业大学 Pavlov dual-mode switching learning memory circuit based on memristor
CN210488595U (en) * 2019-08-25 2020-05-08 湖北大学 Synapse bionic circuit for realizing diversified STDP learning rule based on memristor

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106815636A (en) * 2016-12-30 2017-06-09 华中科技大学 A kind of neuron circuit based on memristor
US20200005130A1 (en) * 2018-07-02 2020-01-02 Kabushiki Kaisha Toshiba Reinforcement learning system
CN110110840A (en) * 2019-04-22 2019-08-09 中国地质大学（武汉） A kind of associative memory emotion recognition circuit based on memristor neural network
CN210488595U (en) * 2019-08-25 2020-05-08 湖北大学 Synapse bionic circuit for realizing diversified STDP learning rule based on memristor
CN110910723A (en) * 2019-12-23 2020-03-24 郑州轻工业大学 Pavlov dual-mode switching learning memory circuit based on memristor

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ZILU WANG ET AL.: "Memristive Circuit Design of Emotional Generation and Evolution Based on Skin-Like Sensory Processor", 《IEEE TRANSACTIONS ON BIOMEDICAL CIRCUITS AND SYSTEMS》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113344194A (en) * 2021-05-31 2021-09-03 郑州轻工业大学 Operational conditioned reflex circuit based on memristor
CN113344194B (en) * 2021-05-31 2024-02-13 郑州轻工业大学 Operational condition reflecting circuit based on memristor
CN113658493A (en) * 2021-08-20 2021-11-16 安徽大学 Reinforced learning bionic circuit architecture for simulating associative memory
CN113658493B (en) * 2021-08-20 2023-05-02 安徽大学 Reinforced learning bionic circuit architecture for simulating associative memory
CN114997387A (en) * 2022-06-28 2022-09-02 郑州轻工业大学 Visual-evoked emotional habituation characteristic neural network circuit
CN117610629A (en) * 2023-12-20 2024-02-27 湖南师范大学 Memristive neural network circuit for inhibiting negative emotion and self-repairing based on bionic mechanism
CN117610629B (en) * 2023-12-20 2024-06-04 湖南师范大学 Memristive neural network circuit for inhibiting negative emotion and self-repairing based on bionic mechanism

Also Published As

Publication number Publication date
CN112651495B (en) 2021-10-12

Similar Documents

Publication Publication Date Title
CN112651495B (en) Neural network circuit for emotion homeostasis adjustment and associative memory
CN110110840B (en) Association memory emotion recognition circuit based on memristor neural network
CN114925818A (en) Memristive layer-level memory neural network circuit for simulating biological learning
US11586887B2 (en) Neural network apparatus
Sun et al. Memristor-based neural network circuit of associative memory with overshadowing and emotion congruent effect
CN114169511B (en) Associative memory circuit and method based on physical memristor
CN115600665A (en) Memristor self-repairing neural network circuit based on VTA-DA neurons
CN110910723B (en) Pavlov dual-mode switching learning memory circuit based on memristor
US20180247179A1 (en) Battery-based neural network weights
CN109376851A (en) The spiking neuron signal generating circuit of bionic system is based on the implementation method of memristor
Bouhadjar et al. Sequence learning in a spiking neuronal network with memristive synapses
CN115456157B (en) Multi-sense interconnection memory network circuit based on memristor
Shin et al. Dynamic range and sensitivity adaptation in a silicon spiking neuron
CN110874629A (en) Structure optimization method of reserve pool network based on excitability and inhibition STDP
JPH07141313A (en) Neural circuit element
CN112953458A (en) Second-order damping emotion circuit based on memristor
CN114970850A (en) Emotional habituation neural network circuit with situation dependence and generalization
CN109255437B (en) A kind of memristor nerve network circuit of flexibly configurable
Jiang et al. Memristive Bionic Memory Circuit Implementation and Its Application in Multisensory Mutual Associative Learning Networks
CN114219082B (en) Improved memristor neuron circuit
CN113408719B (en) Multilevel associative memory circuit based on non-molten state phase change device
CN117610629B (en) Memristive neural network circuit for inhibiting negative emotion and self-repairing based on bionic mechanism
CN118504629A (en) Dynamic random access memory, neuron behavior simulation system and method
Krestinskaya et al. AnalogHTM: Memristive spatial pooler learning with backpropagation
CN116128030A (en) Pavlovian neural network circuit with conditional inhibition and blocking generalization differentiation

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