84 results about "Synaptic device" patented technology

The invention discloses a multi-modal sensory artificial synaptic device based on a framework material and a preparation method and application thereof. The artificial synaptic device comprises a substrate and a channel, wherein the channel is arranged on the surface of the substrate; the channel is an amorphous organic metal framework film which responds to light and gas stimulation; the metal electrode is arranged on the surface of the channel; the metal electrode comprises a source electrode and a drain electrode. Compared with the prior art, the framework material with excellent photoelectric response performance is adopted, and a novel artificial synapse device with multi-mode sensing ability is constructed by fully utilizing the regular pore structure, large specific surface area and good photoresponse characteristic of the MOF material. The two electrodes of the device are connected into electrical test equipment, the conductivity of the amorphous MOF thin film can be changed under the stimulation of optical pulse signals, and finally simulation based on photoresponse synaptic plasticity is achieved. The device is placed in different gas environments, and the conductivity of the MOF film can be changed by capturing and releasing target gas molecules through the porous characteristic of the amorphous MOF material, so that simulation based on gas response synaptic plasticity is realized.

Disclosed is a 3-terminal neuromorphic synaptic device including a substrate, a source electrode and a drain electrode provided on the substrate to be spaced apart from each other, a channel area provided on the substrate to be electrically connected to the source electrode and the drain electrode, between the source electrode and the drain electrode, an ion transport layer provided on the channel area, a gate electrode provided on the ion transport layer, and a voltage application part that applies a gate voltage to the gate electrode. The gate electrode is formed of at least one of an oxide-based material including mobile ions, a chalcogenide-based material including the mobile ions, and a nitride-based material including the mobile ions.

The embodiment of the invention discloses a photoelectric synapse device, the photoelectric synapse device has an optical synapse characteristic in a spectral response range of 520 nm to 2000 nm from visible light to infrared light, the photoelectric synapse device comprises a Van der Waals heterojunction formed by upper graphene / CrOCl / lower graphene, and the Van der Waals heterojunction is formed by the upper graphene / CrOCl / lower graphene. The CrOCl serves as a middle layer and is clamped between the upper-layer graphene and the lower-layer graphene, the number of the upper-layer graphene used for receiving light is 5-10, and the thickness of the CrOCl is set to enable the photoelectric synapse device to be capable of being tunneled when the photoelectric synapse device is irradiated by visible light or infrared light. The photoelectric synaptic device disclosed by the invention efficiently simulates optical synaptic behaviors in a wide spectral range of 520-2000 nm, and has adjustable pulse time sequence dependence plasticity, pulse number dependence plasticity, pulse frequency dependence plasticity and paired pulse promotion. The invention also discloses a preparation method and application of the photoelectric synaptic device.

The invention discloses a double-end photoelectric synapse element based on two-dimensional intrinsic ferroelectricity and TMDs Van der Waals heterojunction as well as a preparation method and application of the double-end photoelectric synapse element, and belongs to the technical field of photoelectric synapses. The element comprises a substrate layer, a channel layer and a packaging layer which are stacked in sequence, the channel layer is formed by stacking a single layer of two-dimensional TMDs and a two-dimensional intrinsic ferroelectric alpha-indium selenide layer; a source electrode is deposited on the TMDs, and a drain electrode is deposited on the alpha-indium selenide layer. The preparation method of the element comprises the following steps: cleaning the substrate layer, growing the single-layer two-dimensional TMDs through a CVD (chemical vapor deposition) method, and stacking the two-dimensional intrinsic ferroelectric alpha-indium selenide layer on the single-layer two-dimensional TMDs to obtain a precursor; the precursor is sequentially subjected to primary annealing, etching and electrode evaporation, and then is subjected to secondary annealing, so that the electrode is obtained. The element adopts photoelectric signal hybrid regulation and control, is used for preparing an artificial brain-like synapse device, and can effectively solve the problems of high power consumption and low dipulse dissimilation index in the prior art.

The embodiment of the invention discloses a ferroelectric domain wall neuronal synaptic device, a regulation and control method thereof and a neural network device. The ferroelectric domain wall nerve synapse device comprises a ferroelectric domain wall unit and an electric field forming unit; each ferroelectric domain wall unit comprises a single crystal substrate layer, a bottom electrode layer and a ferroelectric material layer which are stacked in sequence; and the electric field forming unit is used for forming an equivalent electric field according to the received pulse signal and loading the equivalent electric field on the ferroelectric domain wall unit so as to regulate and control the conductivity value of the ferroelectric domain wall unit. According to the embodiment, the electric field forming unit in the ferroelectric domain wall neural synapse device receives the pulse signal to form the equivalent electric field, and the equivalent electric field is loaded on the ferroelectric domain wall unit to regulate and control the conductivity value of the ferroelectric domain wall unit, so that integration of information storage and neural morphology calculation can be realized; the method has the advantages of being low in environmental influence degree, stable in resistance state, low in energy consumption, high in calculation speed and the like, and has wide application prospects.

The present application belongs to the technical field of nanoelectronic devices, and provides an integrated gas sensing-computing synaptic device and a preparation method therefor. The device comprises: a first functional layer, a barrier layer, a second functional layer, a first metal electrode, and a second metal electrode; the first functional layer is heavily doped silicon; the second functional layer uses CuOx material, such that defect energy levels are formed by interstitial oxygen and copper vacancies; the barrier layer increases contact area between the second functional layer and a test gas; a direct current voltage scan is applied between the first metal electrode and the second metal electrode, to simulate resistive switching behavior of a memristor; the second functional layer reacts with the test gas, to implement a gas sensing function; a constant voltage signal is applied between the first metal electrode and the second metal electrode, and in a gas pulse environment, a current response state exhibits a synaptic response. The present application can implement in-situ sensing and processing of gas information, improving information processing efficiency, and the gas information inference method of the integrated gas sensing-computing device is simpler.

This invention relates to the fields of semiconductor information storage and artificial synaptic devices, specifically to a biological synaptic intrinsically self-assembled topological phase-change memristor and its fabrication method. The fabrication method includes: providing a substrate; depositing a bottom electrode; and depositing SrFeO on the bottom electrode. x The thin film serves as the storage medium layer; the storage medium layer undergoes heat treatment, including in-situ annealing and post-annealing. By controlling the oxygen pressure difference between in-situ and post-annealing, the single-component SrFeO is made more efficient. x An intrinsically self-assembled superlattice structure, characterized by alternating vertical and horizontal orientation regions of oxygen vacancy channels, spontaneously forms within the thin film. This structure restricts the formation sites of conductive filaments, significantly improving the consistency and stability of the device's resistive switching cycle. The process of this invention is simple, requiring no introduction of a second-phase material. The resulting memristor requires no electrical formation, exhibits a high on / off ratio, multi-level storage, and short-term synaptic plasticity, making it applicable to neuromorphic computing chips and artificial intelligence devices.

The application discloses a kind of based on laminated CrOCl auxiliary storage's floating gate structure optoelectronic synapse device and preparation method, belong to semiconductor device technical field.Device includes by lower to upper sequentially arranged substrate, Gr nanosheet, CrOCl nanosheet, h-BN nanosheet and MoS2 Nanosheet, wherein, MoS2 Nanosheet two ends are connected with active electrode and drain electrode, Gr nanosheet, CrOCl nanosheet, h-BN nanosheet and MoS2 Nanosheet are combined by Van der Waals interaction;MoS2 / h-BN / CrOCl / Gr floating gate optoelectronic synapse device of the application can effectively increase double scanning storage window by the secondary adsorption electron capacity of CrOCl, widen multi-level storage capacity, can realize multi-level storage under the joint modulation of electric pulse and light pulse, show excellent electrical writing and optical erasing technology in photoelectric hybrid storage, in addition, under the common modulation of photoelectricity, show good synapse characteristics, including long-term plasticity, short-term plasticity, the recognition accuracy of neural network constructed by MoS2 / h-BN / CrOCl / Gr floating gate device of the application can reach 97% on direction identification data set.

The invention discloses a non-contact handwriting device based on an ultraviolet specific image sensor and a preparation method, and relates to the technical field of image sensing and human-computer interaction. Comprising a glass substrate 1, a transparent thin film transistor array 2 formed on the substrate, a photosensitive material layer 3 formed on the thin film transistor array, and an ultraviolet curing adhesive packaging layer 4 covering the photosensitive material layer. A light-sensitive material is a Bi2S3 material with a wavelength selective light current memory characteristic, generates long-time light current maintenance based on a photo-thermal electric effect in an ultraviolet band, rapidly attenuates in visible to near-infrared bands, and has ultraviolet input selectivity and ambient light anti-interference capability. According to the system, an ultraviolet laser pen is used as an input tool, handwriting patterns which can be kept for a long time are formed on an image sensor array, and image erasing based on a background normalization mechanism is supported. The image sensor array is composed of synaptic-like devices, has a long-term optical memory characteristic, and realizes integration of perception and storage.

The invention discloses a sensing and storage calculation integration method of an electrophysiological sensing device and an artificial synaptic storage calculation device, relates to the technical field of neuromorphic calculation, and aims to solve the problem of separation of wearable health monitoring sensing and storage calculation. The method comprises the following steps: S1, precisely preparing two-dimensional MXene by a liquid phase etching method; s2, constructing an MXene electrophysiological sensing device; s3, constructing an MXene artificial synaptic device; s4, the MXene electrophysiological sensing device in the step S2 and the MXene artificial synapse device in the step S3 are integrated; wherein the step S4 specifically comprises the following steps: S4-1, leading out a source electrode, a grid electrode and a drain electrode of the MXene artificial synapse device through a bonding machine instrument, connecting the grid electrode with the MXene electrophysiological sensor, applying a voltage of 100mV to the source electrode of the MXene artificial synapse device, testing a current change value of a channel material of the MXene artificial synapse device by the drain electrode, and presenting a curve through a mobile terminal, therefore, the human body electrophysiology compression signal can be presented.

This invention discloses a design and operation method for a multimodal optoelectronic synaptic device based on the synergy of asymmetric structure and defects. The method includes constructing a device with a two-terminal asymmetric structure, consisting of a metal layer / functional layer 1 / functional layer 2 / electrode. The core lies in forming a two-terminal asymmetric functional junction (Schottky junction / heterojunction) on both sides of functional layer 1. This invention utilizes the unified yet differentiated response dynamics (migration and recombination) of defects such as oxygen vacancies in the functional layer to light and electrical signals, achieving a response within an order of magnitude of specific encoded light and electrical signals in a single device. This enables weighted fusion of multimodal signals, non-volatile storage, and selective decoupling. This method is independent of specific material systems (including oxide types, electrode types, doping, etc.) and fabrication processes, providing core operational logic for constructing general-purpose neuromorphic hardware capable of performing in-situ sensing and computation.

This invention discloses a fully optically modulated photoelectric synaptic device, its fabrication method, and its applications. This device successfully simulates the synaptic behavior of nerve cells within the ultraviolet to visible light spectrum. Furthermore, the device exhibits positive responses to both ultraviolet and visible light, with an increase in photocurrent. However, when irradiated with ultraviolet light first and then with visible light, the visible light inhibits the photocurrent generated by the ultraviolet light, causing a rapid decrease in photocurrent.

This invention discloses a photonic neural synapse device with a dual-micro-ring structure and a convolutional computation network model. The device includes a front resonant ring, a rear resonant ring, and two coupled phase-change material (PCM) films. The two PCM films are matched and correspond to the front and rear resonant rings, respectively. Synaptic weighting is achieved by controlling the coupled PCM films. The front and rear resonant rings are used to split the input optical signal in the bus waveguide into a 50:50 ratio, and output the difference signal current through a balanced photodetector. The photonic neural synapse of this invention has the advantages of high bandwidth and high speed. Once the synaptic weights are determined, it is a passive device with theoretically zero power consumption. Furthermore, this invention designs a dual-micro-ring structure, which ensures that the modulation of light does not affect the resonant wavelength, improving modulation accuracy, reducing optical transmission loss, and broadening the range of synaptic weights. This invention can be widely applied in the field of micro-nano optoelectronics.

The invention relates to the technical field of optical equipment, and provides a spectrum detection device and a spectrum detection method.The spectrum detection device comprises an artificial synaptic assembly and a signal processing assembly, and the artificial synaptic assembly comprises a plurality of artificial synaptic devices and is used for responding according to different wavelengths of incident light and outputting corresponding induced currents; the signal processing assembly is connected with the plurality of artificial synaptic devices and is used for receiving the induced current and carrying out amplification, analog-to-digital conversion and analysis so as to reconstruct spectral information of the incident light, the artificial synaptic assemblies are directly irradiated by light to be detected, and the corresponding induced current is output by utilizing the artificial synaptic assemblies; the induced current is transmitted to the signal processing assembly, and spectral information of the light to be detected can be output after processing, so that a complex scanning part is not needed, the detection architecture is simplified, the power consumption is reduced, and integrated application is facilitated.

The application discloses a kind of phase-change optical synapse devices and modulation method, preparation method, belong to integrated optoelectronic technology field;Heating layer is set under the heat conduction layer, and the heat conduction layer is transparent heat conduction film, and its thermal conductivity is greater than the thermal conductivity of heating layer, can evenly transmit the current joule heat generated by heating layer to phase-change material layer, reduce the temperature gradient in phase-change material layer plane;On this basis, set the surrounding layer surrounding phase-change material, can be well isolated heating layer and substrate, reduce substrate heat dissipation, improve heating efficiency;Under the dual action of heat conduction layer and surrounding layer, the application can avoid the recrystallization phenomenon that appears in amorphization process, realize the accurate regulation of phase-change material layer phase-change dynamics process, so that phase-change material layer has gradually crystallized ability and gradually amorphous ability, can realize bilateral polymorphic adjustment in larger modulation range, match the LTP and LTD rule of biological synapse.

The invention provides a photoelectron synaptic device based on a Z-type heterojunction and a preparation method and application thereof.The photoelectron synaptic device sequentially comprises a top electrode layer, a Z-type heterojunction functional layer and a bottom electrode layer from top to bottom, and the Z-type heterojunction functional layer is formed by tightly contacting a first semiconductor material layer and a second semiconductor material layer; and energy bands of the first semiconductor material layer and the second semiconductor material layer are staggered to form a Z-shaped energy band. The photoelectron synaptic device is simple in structure, the Z-type heterojunction functional layer is designed, and a plurality of synaptic behaviors are simulated by using a unique carrier recombination mechanism of the Z-type heterojunction, so that bidirectional synaptic function simulation of inhibition / excitation under the same photostimulation can be realized only through voltage switching, and the photoelectron synaptic device is applicable to various synaptic functions. And a core thought is provided for the design of a multifunctional integrated photoelectron synaptic device.

The invention belongs to the technical field of electric bionic synaptic devices, and particularly relates to a photon memristor for audio-visual dual-mode identification, a preparation method, a storage pool mapping method and application. The memristor comprises a bottom electrode, a hole layer, a light absorption layer, a passivation layer and a top electrode which are sequentially arranged from bottom to top, the hole layer is a ZnO thin film, the light absorption layer is a Cs2AgBiBr6 thin film, and the ZnO thin film and the Cs2AgBiBr6 thin film form a heterojunction; the passivation layer is a PMMA film. According to the photon memristor provided by the invention, the conductivity value can be regulated and controlled through ultraviolet pulse programming, and the conductivity state can provide abundant storage state space for physical storage pool calculation. In letter and audio recognition, the accuracy rates of 91.7% and 95.0% can be respectively realized based on physical storage pool calculation of the photon memristor. The application of the method can greatly improve the application of the photoelectric bionic synaptic device in audio-visual multi-mode identification, and provides a referable device instance for the implementation of physical storage pool calculation.

The application discloses a two-end vertical structure organic electrochemical biomimetic synapse device and a preparation method and a biomimetic neural morphological chip thereof. The device comprises a bottom electrode, a first organic mixed ion-electron conductor layer, a first ion gel layer and a top electrode which are sequentially stacked. The first ion gel layer is configured to provide anions. The first organic mixed ion-electron conductor layer is configured to transport anions and electrons. The top electrode is configured to apply a bias voltage to drive the migration of anions. In terms of structure, the four-layer stacked structure realizes ion path innovation, and the vertical channel sharply reduces the delay time to the microsecond level. In terms of device mechanism, the low reduction potential of the organic ion-electron mixed conductor and the depletion of the de-doped carriers are utilized, the current spontaneously reverses under the action of a certain voltage, and the simulation of artificial synapse sensitization and habituation is realized at the single device level.

The invention relates to the technical field of electronic information, and discloses a heterojunction artificial synapse device with multi-wavelength response and bionic injury perception, which comprises a substrate, a silicon dioxide layer, a silicon-rich silicon nitride layer, a molybdenum disulfide layer and an electrode layer, the silicon dioxide layer covers the surface of the substrate, the silicon-rich silicon nitride layer is deposited on the side, away from the substrate, of the silicon dioxide layer, the molybdenum disulfide layer is transferred to the side, away from the silicon dioxide layer, of the silicon-rich silicon nitride layer, and the electrode layer is arranged on the side, away from the silicon-rich silicon nitride layer, of the molybdenum disulfide layer; and the substrate is a p-type silicon substrate. By constructing a molybdenum disulfide and silicon-rich silicon nitride heterojunction structure and optimizing material parameters and a preparation process of each layer, efficient response of the device to 350nm-380nm short-wave-band light and 510nm-550nm long-wave-band light is realized, light current generated under short-wave-band light stimulation is larger than light current generated under long-wave-band light stimulation, the multi-wavelength visual information perception requirement is met, and the device has a wide application prospect. And hardware support is provided for optical signal identification and processing in a complex environment.

The invention belongs to the technical field of photoelectric detection, and particularly relates to a near-infrared photoelectric synapse device and a preparation method thereof. The near-infrared photoelectric synapse device comprises a conductive substrate, a transition metal sulfide layer, a perovskite active layer and an electrode layer which are stacked from bottom to top, wherein when the thickness of the transition metal sulfide layer is gradually increased within a preset thickness range, a high-density interface capture state is formed between the transition metal sulfide layer and the perovskite active layer, so that the double-pulse facilitation index of the near-infrared photoelectric synapse device is gradually increased under the stimulation of near-infrared pulses; therefore, photoelectric response and short-time synaptic plasticity are realized in a near-infrared band. The photoelectric synaptic device designed in the scheme can sense infrared signals, the working wave band range of the device is widened, and meanwhile, the photoelectric synaptic device has a high double-pulse simplification index, namely, the device can show a low-energy-consumption strong synaptic behavior in a near-infrared wave band, so that the photoelectric synaptic device can be used for constructing a brain-like vision and neuromorphic computing system.

The embodiment of the invention discloses a light-operated artificial synaptic device and a preparation method thereof. The artificial synaptic device comprises a substrate, a light-emitting layer and a synaptic function layer which are sequentially stacked in the thickness direction: the light-emitting layer is arranged on the substrate, and the light-emitting surface of the light-emitting layer faces the synaptic function layer; and the synaptic function layer is used for responding to the light signal emitted by the light-emitting layer and generating a conductivity change for simulating a biological synaptic behavior. According to the light-operated artificial synapse device provided by the invention, the light-emitting layer is integrated into the artificial synapse device, so that light signals do not need to be propagated for a long distance and can directly act on the synapse function layer in a directional manner, and the problem of low light coupling efficiency of the existing light-operated synapse is further solved.

This application belongs to the interdisciplinary fields of semiconductor optoelectronic devices, multiferroic materials, and neuromorphic computing. Specifically, it discloses an in-situ differential optoelectronic synaptic device and its in-memory computing method. The method involves grounding the first and second electrodes, applying a voltage to the bottom electrode to change the polarization direction of the ferroelectric domains in the bismuth ferrite thin film, and using the polarization direction of the ferroelectric domains as the weight symbol in the neural weights to set the non-volatile synaptic weights. The voltage on the bottom electrode is then removed or reduced, and a light spot is used to illuminate the bismuth ferrite thin film. By adjusting the position of the light spot, the differential current between the first and second electrodes is read and used as the weight value in the neural weights, thus realizing the read operation. This application achieves integrated storage and computing.

The application relates to a preparation method of a bismuth selenide film with adjustable defect concentration and application of the bismuth selenide film to a photoelectric synaptic device, and belongs to the technical field of two-dimensional photoelectric materials and photoelectric synaptic device preparation. The specific scheme is as follows: a Bi2Se3 film is prepared by adopting chemical vapor deposition, secondary treatment is conducted on the prepared Bi2Se3 film by in-situ heat treatment for avoiding air contact and keeping the original atmosphere to obtain a bismuth selenide film with adjustable defect concentration. The photoelectric synaptic device of the prepared bismuth selenide film is of an MSM structure, is simple in structure and has integration potential; the photoelectric response of the Bi2Se3 film photoelectric synaptic device has obvious basic synaptic behaviors of long-term potentiation, short-term potentiation and pulse enhancement, and can be applied to the field of artificial intelligence.

The invention relates to a charge storage type synaptic device for deep neural network training and a driving method thereof. The synaptic device disclosed by the invention comprises a weighting capacitor which is provided with a first terminal and a second terminal and is used for storing a voltage corresponding to a weighting value; the four control transistors are used for changing the voltage or the weighted value of the capacitor; a first output transistor including a first gate terminal coupled to a first terminal of the capacitor and outputting a first drain current in accordance with a voltage applied to the first gate terminal; and a second output transistor including a second gate terminal coupled to a second terminal of the capacitor and outputting a second drain current in accordance with a voltage applied to the second gate terminal, the weighting value being dependent on a difference between the first drain current and the second drain current.

The application provides a wide-area refresh rate display driving system based on a neuromorphic device and a working method thereof, and belongs to the technical field of display, and comprises an input module, a control module and a light-emitting module; the control module comprises an enabling control module and a brightness control module which are connected with each other, the input module is connected with the enabling control module, and the light-emitting module is connected with the brightness control module; the input module splices and inputs an input enabling signal and an input data signal into the control module; the enabling control module adopts an artificial neuron device, and a single artificial neuron device is switched between on and off states according to the input enabling signal; the brightness control module adopts an artificial synapse device, and a single artificial synapse device is transformed among a plurality of different conductance states according to the input data signal; and the light-emitting module performs light-emitting display under the regulation and control of the brightness control module. The application can not only realize pixel-level control through an artificial neuron, but also realize wide-area refresh rate through an artificial synapse.

The invention belongs to the technical field of microelectronics, and particularly relates to an artificial synaptic device based on a vertical nanowire array and a preparation method and application of the artificial synaptic device. The synaptic device comprises a substrate, a bottom electrode layer, an insulating template, a vertex electrode layer and a nanowire array. And the bottom electrode layer, the insulating template and the top electrode layer form a sandwich structure. The insulating template internally comprises a large number of vertically through nanopores, and the nanowire array is filled in the nanopores and is formed by a chalcogenide semiconductor material or a topological insulator material with rich defect states; one end of the nanowire array is electrically contacted with the bottom electrode layer, and a barrier layer is arranged between the other end of the nanowire array and the top electrode layer. Reversible switching of the device between a volatile threshold switching mode and a non-volatile resistive random access memory mode can be realized by adjusting the magnitude and the direction of a scanning voltage and a limiting current applied between the top electrode and the bottom electrode; the device overcomes the defects of a traditional device in the aspects of functional flexibility and structural reliability.

A programming circuit includes a time difference converter circuit and a pulse generator circuit. The converter circuit comprises first and second circuits configured to generate, based on a clock signal, first and second signals corresponding to first and second pulses from first and second neuron devices in a neural network, to output a time difference signal corresponding to a time difference between the first pulse and the second pulse. The neural network further comprises a synapse device coupled between the first neuron device and the second neuron device. The pulse generator circuit is configured to output a program voltage corresponding to the time difference signal to program a weight value in the synapse device with the program voltage.

Provided is a method of manufacturing a synaptic device. The method includes forming a first electrode, forming a synaptic mimic layer including a hole transport layer and an electron transport layer on the first electrode, and forming a second electrode on the synaptic mimic layer, wherein the forming of the synaptic mimic layer includes forming the electron transport layer on the hole transport layer through a solution process.

The application discloses a method for neural network online training and reasoning based on a CMOS transistor hybrid memory circuit, which adopts a CMOS transistor synapse device and a CMOS capacitor to form a synapse unit, replaces a traditional memristor and a ferroelectric capacitor, the writing durability of the CMOS transistor synapse device is far superior to that of the traditional memristor, the reading operation of the CMOS capacitor is non-destructive, the life loss caused by data rewriting is avoided, the whole process is realized based on a standard CMOS process, is fully compatible with an existing integrated circuit manufacturing process, and the hardware production cost is reduced, only analog weights are called in the reasoning stage to perform operation, and there is no additional data transmission; in the weight transmission stage, an external DAC is not needed, digital-to-analog conversion is realized by using a circuit parasitic parameter, and the energy consumption of an edge device is greatly reduced.

The invention belongs to the technical field of memristive synaptic devices, and particularly relates to a memristive synaptic device of a two-dimensional / three-dimensional perovskite thin film regulated and controlled through cation cover plate drying and annealing and a preparation method of the memristive synaptic device. Fluorinated functionalized two-dimensional perovskite (4, 4-DFHHA) 2PbI4 and three-dimensional perovskite are combined to serve as a resistive layer, and the ITO / three-dimensional perovskite / two-dimensional perovskite / Ag structure memristor is prepared by combining one-step spin coating and cation cover plate drying. According to the method, the defect density is reduced and the carrier mobility is improved through the cation cover plate annealing process, the prepared memristor has excellent resistance switching characteristics and environmental stability, the switch ratio reaches 104, the cycle durability exceeds 103 times, and a new scheme is provided for artificial synapse application.